feat: 实现统一模型 ID 机制

实现统一模型 ID 格式 (provider_id/model_name)，支持跨协议模型标识和 Smart Passthrough。

核心变更：
- 新增 pkg/modelid 包：解析、格式化、校验统一模型 ID
- 数据库迁移：models 表使用 UUID 主键 + UNIQUE(provider_id, model_name) 约束
- Repository 层：FindByProviderAndModelName、ListEnabled 方法
- Service 层：联合唯一校验、provider ID 字符集校验
- Conversion 层：ExtractModelName、RewriteRequestModelName/RewriteResponseModelName 方法
- Handler 层：统一模型 ID 路由、Smart Passthrough、Models API 本地聚合
- 新增 error-responses、unified-model-id 规范

测试覆盖：
- 单元测试：modelid、conversion、handler、service、repository
- 集成测试：统一模型 ID 路由、Smart Passthrough 保真性、跨协议转换
- 迁移测试：UUID 主键、UNIQUE 约束、级联删除

OpenSpec：
- 归档 unified-model-id 变更到 archive/2026-04-21-unified-model-id
- 同步 11 个 delta specs 到 main specs
- 新增 error-responses、unified-model-id 规范文件

.gitignore

@@ -404,4 +404,5 @@ cython_debug/
 openspec/changes/archive
 temp
 .agents
-skills-lock.json
+skills-lock.json
+.worktrees

README.md

@@ -38,10 +38,11 @@ nex/
 ## 功能特性
 
 - **双协议支持**：同时支持 OpenAI 和 Anthropic 协议
-- **透明代理**：对 OpenAI 兼容供应商透传请求
+- **统一模型 ID**：`provider_id/model_name` 格式全局唯一标识模型（如 `openai/gpt-4`）
+- **透明代理**：对 OpenAI 兼容供应商 Smart Passthrough，最小化改写保持参数保真
 - **流式响应**：完整支持 SSE 流式传输
 - **Function Calling**：支持工具调用（Tools）
-- **多供应商管理**：配置和管理多个供应商
+- **多供应商管理**：配置和管理多个供应商（供应商 ID 仅限字母、数字、下划线）
 - **用量统计**：按供应商、模型、日期统计请求数量
 - **Web 配置界面**：提供供应商和模型配置管理
 
@@ -99,23 +100,27 @@ bun dev
 
 ### 代理接口（对外部应用）
 
+代理接口统一使用 `provider_id/model_name` 格式的模型 ID（如 `openai/gpt-4`）。同协议请求走 Smart Passthrough，最小化 JSON 改写保持参数保真；跨协议请求走完整 decode/encode 转换。
+
 - `POST /v1/chat/completions` - OpenAI Chat Completions API
 - `POST /v1/messages` - Anthropic Messages API
+- `GET /v1/models` - 模型列表（本地数据库聚合，不请求上游）
+- `GET /v1/models/{provider_id}/{model_name}` - 模型详情（本地数据库查询）
 
 ### 管理接口（对前端）
 
 #### 供应商管理
 - `GET /api/providers` - 列出所有供应商
-- `POST /api/providers` - 创建供应商
+- `POST /api/providers` - 创建供应商（`id` 仅限字母、数字、下划线，长度 1-64）
 - `GET /api/providers/:id` - 获取供应商
-- `PUT /api/providers/:id` - 更新供应商
+- `PUT /api/providers/:id` - 更新供应商（`id` 不可修改）
 - `DELETE /api/providers/:id` - 删除供应商
 
 #### 模型管理
 - `GET /api/models` - 列出模型（支持 `?provider_id=xxx` 过滤）
-- `POST /api/models` - 创建模型
+- `POST /api/models` - 创建模型（`id` 由系统自动生成 UUID，`provider_id` + `model_name` 联合唯一）
-- `GET /api/models/:id` - 获取模型
+- `GET /api/models/:id` - 获取模型（响应含 `unified_id` 字段，格式 `provider_id/model_name`）
-- `PUT /api/models/:id` - 更新模型
+- `PUT /api/models/:id` - 更新模型（不可修改 `id`）
 - `DELETE /api/models/:id` - 删除模型
 
 #### 统计查询

backend/README.md

@@ -108,12 +108,13 @@ backend/
 │   │   ├── logger.go
 │   │   ├── rotate.go
 │   │   └── context.go
+│   ├── modelid/                 # 统一模型 ID 工具包
+│   │   ├── model_id.go
+│   │   └── model_id_test.go
 │   └── validator/               # 验证器
 │       └── validator.go
 ├── migrations/                  # 数据库迁移
-│   ├── 20260401000001_initial_schema.sql
+│   └── 20260421000001_initial_schema.sql
-│   ├── 20260401000002_add_indexes.sql
-│   └── 20260419000001_add_provider_protocol.sql
 ├── tests/                       # 集成测试
 │   ├── helpers.go
 │   └── integration/
@@ -292,6 +293,8 @@ GET  /anthropic/v1/models
 
 **协议转换**：网关支持任意协议间的双向转换。客户端使用 OpenAI 协议请求，上游供应商可以是 Anthropic 协议（反之亦然）。同协议时自动透传，零序列化开销。
 
+**统一模型 ID**：代理请求中的 `model` 字段使用 `provider_id/model_name` 格式（如 `openai/gpt-4`），网关据此路由到对应供应商。同协议时自动改写为上游 `model_name`，跨协议时通过全量转换处理。
+
 ### 管理接口
 
 #### 供应商管理
@@ -324,14 +327,30 @@ GET  /anthropic/v1/models
 - `PUT /api/models/:id` - 更新模型
 - `DELETE /api/models/:id` - 删除模型
 
+**创建请求**（id 由系统自动生成 UUID）：
+
 ```json
 {
-  "id": "gpt-4",
   "provider_id": "openai",
   "model_name": "gpt-4"
 }
 ```
 
+**响应示例**：
+
+```json
+{
+  "id": "550e8400-e29b-41d4-a716-446655440000",
+  "provider_id": "openai",
+  "model_name": "gpt-4",
+  "unified_id": "openai/gpt-4",
+  "enabled": true,
+  "created_at": "2026-04-21T00:00:00Z"
+}
+```
+
+**统一模型 ID**：`unified_id` 字段为 `provider_id/model_name` 格式，用于代理请求的 `model` 参数。
+
 #### 统计查询
 
 - `GET /api/stats` - 查询统计

backend/cmd/server/main.go

@@ -68,7 +68,7 @@ func main() {
 	statsRepo := repository.NewStatsRepository(db)
 
 	// 5. 初始化 service 层
-	providerService := service.NewProviderService(providerRepo)
+	providerService := service.NewProviderService(providerRepo, modelRepo)
 	modelService := service.NewModelService(modelRepo, providerRepo)
 	routingService := service.NewRoutingService(modelRepo, providerRepo)
 	statsService := service.NewStatsService(statsRepo)

backend/internal/config/models.go

@@ -17,11 +17,11 @@ type Provider struct {
 	Models    []Model        `gorm:"foreignKey:ProviderID;constraint:OnDelete:CASCADE" json:"models,omitempty"`
 }
 
-// Model 模型配置
+// Model 模型配置（id 为 UUID 自动生成，UNIQUE(provider_id, model_name)）
 type Model struct {
 	ID         string    `gorm:"primaryKey" json:"id"`
-	ProviderID string    `gorm:"not null;index" json:"provider_id"`
+	ProviderID string    `gorm:"not null;index;uniqueIndex:idx_provider_model" json:"provider_id"`
-	ModelName  string    `gorm:"not null;index" json:"model_name"`
+	ModelName  string    `gorm:"not null;index;uniqueIndex:idx_provider_model" json:"model_name"`
 	Enabled    bool      `gorm:"default:true" json:"enabled"`
 	CreatedAt  time.Time `json:"created_at"`
 }

backend/internal/conversion/adapter.go

@@ -40,6 +40,12 @@ type ProtocolAdapter interface {
 	EncodeRerankRequest(req *canonical.CanonicalRerankRequest, provider *TargetProvider) ([]byte, error)
 	DecodeRerankResponse(raw []byte) (*canonical.CanonicalRerankResponse, error)
 	EncodeRerankResponse(resp *canonical.CanonicalRerankResponse) ([]byte, error)
+
+	// 统一模型 ID 相关方法
+	ExtractUnifiedModelID(nativePath string) (string, error)
+	ExtractModelName(body []byte, ifaceType InterfaceType) (string, error)
+	RewriteRequestModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error)
+	RewriteResponseModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error)
 }
 
 // AdapterRegistry 适配器注册表接口

backend/internal/conversion/anthropic/adapter.go

@@ -2,6 +2,7 @@ package anthropic
 
 import (
 	"encoding/json"
+	"fmt"
 	"strings"
 
 	"nex/backend/internal/conversion"
@@ -39,13 +40,13 @@ func (a *Adapter) DetectInterfaceType(nativePath string) conversion.InterfaceTyp
 	}
 }
 
-// isModelInfoPath 判断是否为模型详情路径（/v1/models/{id}）
+// isModelInfoPath 判断是否为模型详情路径（/v1/models/{id}，允许 id 含 /）
 func isModelInfoPath(path string) bool {
 	if !strings.HasPrefix(path, "/v1/models/") {
 		return false
 	}
 	suffix := path[len("/v1/models/"):]
-	return suffix != "" && !strings.Contains(suffix, "/")
+	return suffix != ""
 }
 
 // BuildUrl 根据接口类型构建 URL
@@ -203,3 +204,74 @@ func (a *Adapter) DecodeRerankResponse(raw []byte) (*canonical.CanonicalRerankRe
 func (a *Adapter) EncodeRerankResponse(resp *canonical.CanonicalRerankResponse) ([]byte, error) {
 	return nil, conversion.NewConversionError(conversion.ErrorCodeInterfaceNotSupported, "Anthropic 不支持 Rerank 接口")
 }
+
+// ExtractUnifiedModelID 从路径中提取统一模型 ID（/v1/models/{provider_id}/{model_name}）
+func (a *Adapter) ExtractUnifiedModelID(nativePath string) (string, error) {
+	if !strings.HasPrefix(nativePath, "/v1/models/") {
+		return "", fmt.Errorf("不是模型详情路径: %s", nativePath)
+	}
+	suffix := nativePath[len("/v1/models/"):]
+	if suffix == "" {
+		return "", fmt.Errorf("路径缺少模型 ID")
+	}
+	return suffix, nil
+}
+
+// locateModelFieldInRequest 定位请求体中 model 字段的值并提供改写函数
+func locateModelFieldInRequest(body []byte, ifaceType conversion.InterfaceType) (string, func(string) ([]byte, error), error) {
+	var m map[string]json.RawMessage
+	if err := json.Unmarshal(body, &m); err != nil {
+		return "", nil, err
+	}
+
+	switch ifaceType {
+	case conversion.InterfaceTypeChat:
+		raw, exists := m["model"]
+		if !exists {
+			return "", nil, fmt.Errorf("请求体中缺少 model 字段")
+		}
+		var current string
+		if err := json.Unmarshal(raw, &current); err != nil {
+			return "", nil, err
+		}
+		rewriteFunc := func(newModel string) ([]byte, error) {
+			m["model"], _ = json.Marshal(newModel)
+			return json.Marshal(m)
+		}
+		return current, rewriteFunc, nil
+	default:
+		return "", nil, fmt.Errorf("不支持的接口类型: %s", ifaceType)
+	}
+}
+
+// ExtractModelName 从请求体中提取 model 值
+func (a *Adapter) ExtractModelName(body []byte, ifaceType conversion.InterfaceType) (string, error) {
+	model, _, err := locateModelFieldInRequest(body, ifaceType)
+	return model, err
+}
+
+// RewriteRequestModelName 最小化改写请求体中的 model 字段
+func (a *Adapter) RewriteRequestModelName(body []byte, newModel string, ifaceType conversion.InterfaceType) ([]byte, error) {
+	_, rewriteFunc, err := locateModelFieldInRequest(body, ifaceType)
+	if err != nil {
+		return nil, err
+	}
+	return rewriteFunc(newModel)
+}
+
+// RewriteResponseModelName 最小化改写响应体中的 model 字段
+func (a *Adapter) RewriteResponseModelName(body []byte, newModel string, ifaceType conversion.InterfaceType) ([]byte, error) {
+	var m map[string]json.RawMessage
+	if err := json.Unmarshal(body, &m); err != nil {
+		return nil, err
+	}
+
+	switch ifaceType {
+	case conversion.InterfaceTypeChat:
+		// Chat 响应必须有 model 字段，存在则改写，不存在则添加
+		m["model"], _ = json.Marshal(newModel)
+		return json.Marshal(m)
+	default:
+		return body, nil
+	}
+}

backend/internal/conversion/anthropic/adapter_unified_test.go

@@ -0,0 +1,263 @@
+package anthropic
+
+import (
+	"encoding/json"
+	"testing"
+
+	"nex/backend/internal/conversion"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// ---------------------------------------------------------------------------
+// ExtractUnifiedModelID
+// ---------------------------------------------------------------------------
+
+func TestExtractUnifiedModelID(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("standard_path", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/anthropic/claude-3")
+		require.NoError(t, err)
+		assert.Equal(t, "anthropic/claude-3", id)
+	})
+
+	t.Run("multi_segment_path", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/some/deep/nested/model")
+		require.NoError(t, err)
+		assert.Equal(t, "some/deep/nested/model", id)
+	})
+
+	t.Run("single_segment", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/claude-3")
+		require.NoError(t, err)
+		assert.Equal(t, "claude-3", id)
+	})
+
+	t.Run("non_model_path", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/messages")
+		require.Error(t, err)
+	})
+
+	t.Run("empty_suffix", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/models/")
+		require.Error(t, err)
+	})
+
+	t.Run("models_list_no_slash", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/models")
+		require.Error(t, err)
+	})
+
+	t.Run("unrelated_path", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/other")
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// ExtractModelName (Chat only for Anthropic)
+// ---------------------------------------------------------------------------
+
+func TestExtractModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3","messages":[]}`)
+		model, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "anthropic/claude-3", model)
+	})
+
+	t.Run("chat_with_max_tokens", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3-opus","messages":[{"role":"user","content":"hi"}],"max_tokens":1024}`)
+		model, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "anthropic/claude-3-opus", model)
+	})
+
+	t.Run("no_model_field", func(t *testing.T) {
+		body := []byte(`{"messages":[]}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("unsupported_interface_type_embedding", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3"}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeEmbeddings)
+		require.Error(t, err)
+	})
+
+	t.Run("unsupported_interface_type_rerank", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3"}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeRerank)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// RewriteRequestModelName (Chat only for Anthropic)
+// ---------------------------------------------------------------------------
+
+func TestRewriteRequestModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3","messages":[]}`)
+		rewritten, err := a.RewriteRequestModelName(body, "claude-3", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "claude-3", m["model"])
+
+		msgs, ok := m["messages"]
+		require.True(t, ok)
+		msgsArr, ok := msgs.([]interface{})
+		require.True(t, ok)
+		assert.Len(t, msgsArr, 0)
+	})
+
+	t.Run("preserves_unknown_fields", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3","max_tokens":1024,"temperature":0.7}`)
+		rewritten, err := a.RewriteRequestModelName(body, "claude-3", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "claude-3", m["model"])
+		assert.Equal(t, 0.7, m["temperature"])
+
+		// max_tokens is encoded as float in JSON numbers
+		maxTokens, ok := m["max_tokens"]
+		require.True(t, ok)
+		assert.Equal(t, float64(1024), maxTokens)
+	})
+
+	t.Run("no_model_field", func(t *testing.T) {
+		body := []byte(`{"messages":[]}`)
+		_, err := a.RewriteRequestModelName(body, "claude-3", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.RewriteRequestModelName(body, "claude-3", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("unsupported_interface_type", func(t *testing.T) {
+		body := []byte(`{"model":"anthropic/claude-3"}`)
+		_, err := a.RewriteRequestModelName(body, "claude-3", conversion.InterfaceTypeEmbeddings)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// RewriteResponseModelName (Chat only for Anthropic)
+// ---------------------------------------------------------------------------
+
+func TestRewriteResponseModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat_existing_model", func(t *testing.T) {
+		body := []byte(`{"model":"claude-3","content":[],"stop_reason":"end_turn"}`)
+		rewritten, err := a.RewriteResponseModelName(body, "anthropic/claude-3", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "anthropic/claude-3", m["model"])
+
+		// other fields preserved
+		_, hasContent := m["content"]
+		assert.True(t, hasContent)
+		assert.Equal(t, "end_turn", m["stop_reason"])
+	})
+
+	t.Run("chat_without_model_field_adds_it", func(t *testing.T) {
+		body := []byte(`{"content":[],"stop_reason":"end_turn"}`)
+		rewritten, err := a.RewriteResponseModelName(body, "anthropic/claude-3", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "anthropic/claude-3", m["model"])
+	})
+
+	t.Run("passthrough_returns_body_unchanged", func(t *testing.T) {
+		body := []byte(`{"model":"claude-3"}`)
+		rewritten, err := a.RewriteResponseModelName(body, "anthropic/claude-3", conversion.InterfaceTypePassthrough)
+		require.NoError(t, err)
+		assert.Equal(t, string(body), string(rewritten))
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.RewriteResponseModelName(body, "anthropic/claude-3", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// ExtractModelName and RewriteRequest consistency
+// ---------------------------------------------------------------------------
+
+func TestExtractModelNameAndRewriteRequestConsistency(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat_round_trip", func(t *testing.T) {
+		original := []byte(`{"model":"anthropic/claude-3","messages":[],"max_tokens":1024}`)
+
+		// Extract the unified model ID from the body
+		extracted, err := a.ExtractModelName(original, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "anthropic/claude-3", extracted)
+
+		// Rewrite to the native model name
+		rewritten, err := a.RewriteRequestModelName(original, "claude-3", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		// Extract again from the rewritten body to verify the same location was targeted
+		afterRewrite, err := a.ExtractModelName(rewritten, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "claude-3", afterRewrite)
+
+		// Verify other fields are preserved
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, float64(1024), m["max_tokens"])
+	})
+}
+
+// ---------------------------------------------------------------------------
+// isModelInfoPath (additional unified model ID cases)
+// ---------------------------------------------------------------------------
+
+func TestIsModelInfoPath_UnifiedModelID(t *testing.T) {
+	tests := []struct {
+		name     string
+		path     string
+		expected bool
+	}{
+		{"simple_model_id", "/v1/models/claude-3", true},
+		{"unified_model_id_with_slash", "/v1/models/anthropic/claude-3", true},
+		{"models_list", "/v1/models", false},
+		{"models_list_trailing_slash", "/v1/models/", false},
+		{"messages_path", "/v1/messages", false},
+		{"deeply_nested", "/v1/models/org/workspace/claude-3-opus", true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			assert.Equal(t, tt.expected, isModelInfoPath(tt.path))
+		})
+	}
+}

backend/internal/conversion/engine.go

@@ -79,11 +79,29 @@ func (e *ConversionEngine) ConvertHttpRequest(spec HTTPRequestSpec, clientProtoc
 		if err != nil {
 			return nil, err
 		}
+
+		// Smart Passthrough: 同协议时最小化改写 model 字段
+		interfaceType := providerAdapter.DetectInterfaceType(nativePath)
+		rewrittenBody := spec.Body
+
+		// 对于 Chat/Embedding/Rerank 接口，改写请求体中的 model 字段
+		if interfaceType == InterfaceTypeChat || interfaceType == InterfaceTypeEmbeddings || interfaceType == InterfaceTypeRerank {
+			if len(spec.Body) > 0 && provider.ModelName != "" {
+				rewrittenBody, err = providerAdapter.RewriteRequestModelName(spec.Body, provider.ModelName, interfaceType)
+				if err != nil {
+					e.logger.Warn("Smart Passthrough 改写请求失败，使用原始请求体",
+						zap.String("error", err.Error()),
+						zap.String("interface", string(interfaceType)))
+					rewrittenBody = spec.Body
+				}
+			}
+		}
+
 		return &HTTPRequestSpec{
 			URL:     provider.BaseURL + nativePath,
 			Method:  spec.Method,
 			Headers: providerAdapter.BuildHeaders(provider),
-			Body:    spec.Body,
+			Body:    rewrittenBody,
 		}, nil
 	}
 
@@ -112,9 +130,30 @@ func (e *ConversionEngine) ConvertHttpRequest(spec HTTPRequestSpec, clientProtoc
 	}, nil
 }
 
-// ConvertHttpResponse 转换 HTTP 响应
+// ConvertHttpResponse 转换 HTTP 响应，modelOverride 用于跨协议场景覆写 model 字段
-func (e *ConversionEngine) ConvertHttpResponse(spec HTTPResponseSpec, clientProtocol, providerProtocol string, interfaceType InterfaceType) (*HTTPResponseSpec, error) {
+func (e *ConversionEngine) ConvertHttpResponse(spec HTTPResponseSpec, clientProtocol, providerProtocol string, interfaceType InterfaceType, modelOverride string) (*HTTPResponseSpec, error) {
 	if e.IsPassthrough(clientProtocol, providerProtocol) {
+		// Smart Passthrough: 同协议时最小化改写 model 字段
+		if modelOverride != "" && len(spec.Body) > 0 {
+			adapter, err := e.registry.Get(clientProtocol)
+			if err != nil {
+				return &spec, nil
+			}
+
+			rewrittenBody, err := adapter.RewriteResponseModelName(spec.Body, modelOverride, interfaceType)
+			if err != nil {
+				e.logger.Warn("Smart Passthrough 改写响应失败，使用原始响应体",
+					zap.String("error", err.Error()),
+					zap.String("interface", string(interfaceType)))
+				return &spec, nil
+			}
+
+			return &HTTPResponseSpec{
+				StatusCode: spec.StatusCode,
+				Headers:    spec.Headers,
+				Body:       rewrittenBody,
+			}, nil
+		}
 		return &spec, nil
 	}
 
@@ -127,7 +166,7 @@ func (e *ConversionEngine) ConvertHttpResponse(spec HTTPResponseSpec, clientProt
 		return nil, err
 	}
 
-	convertedBody, err := e.convertResponseBody(interfaceType, clientAdapter, providerAdapter, spec.Body)
+	convertedBody, err := e.convertResponseBody(interfaceType, clientAdapter, providerAdapter, spec.Body, modelOverride)
 	if err != nil {
 		return nil, err
 	}
@@ -139,9 +178,17 @@ func (e *ConversionEngine) ConvertHttpResponse(spec HTTPResponseSpec, clientProt
 	}, nil
 }
 
-// CreateStreamConverter 创建流式转换器
+// CreateStreamConverter 创建流式转换器，modelOverride 用于跨协议场景覆写 model 字段
-func (e *ConversionEngine) CreateStreamConverter(clientProtocol, providerProtocol string) (StreamConverter, error) {
+func (e *ConversionEngine) CreateStreamConverter(clientProtocol, providerProtocol string, modelOverride string, interfaceType InterfaceType) (StreamConverter, error) {
 	if e.IsPassthrough(clientProtocol, providerProtocol) {
+		// Smart Passthrough: 同协议流式场景需要逐 chunk 改写 model 字段
+		if modelOverride != "" {
+			adapter, err := e.registry.Get(clientProtocol)
+			if err != nil {
+				return NewPassthroughStreamConverter(), nil
+			}
+			return NewSmartPassthroughStreamConverter(adapter, modelOverride, interfaceType), nil
+		}
 		return NewPassthroughStreamConverter(), nil
 	}
 
@@ -167,6 +214,7 @@ func (e *ConversionEngine) CreateStreamConverter(clientProtocol, providerProtoco
 		ctx,
 		clientProtocol,
 		providerProtocol,
+		modelOverride,
 	), nil
 }
 
@@ -192,11 +240,11 @@ func (e *ConversionEngine) convertBody(interfaceType InterfaceType, clientAdapte
 	}
 }
 
-// convertResponseBody 转换响应体
+// convertResponseBody 转换响应体，modelOverride 非空时在 canonical 层面覆写 Model 字段
-func (e *ConversionEngine) convertResponseBody(interfaceType InterfaceType, clientAdapter, providerAdapter ProtocolAdapter, body []byte) ([]byte, error) {
+func (e *ConversionEngine) convertResponseBody(interfaceType InterfaceType, clientAdapter, providerAdapter ProtocolAdapter, body []byte, modelOverride string) ([]byte, error) {
 	switch interfaceType {
 	case InterfaceTypeChat:
-		return e.convertChatResponseBody(clientAdapter, providerAdapter, body)
+		return e.convertChatResponseBody(clientAdapter, providerAdapter, body, modelOverride)
 	case InterfaceTypeModels:
 		if !clientAdapter.SupportsInterface(InterfaceTypeModels) || !providerAdapter.SupportsInterface(InterfaceTypeModels) {
 			return body, nil
@@ -211,12 +259,12 @@ func (e *ConversionEngine) convertResponseBody(interfaceType InterfaceType, clie
 		if !clientAdapter.SupportsInterface(InterfaceTypeEmbeddings) || !providerAdapter.SupportsInterface(InterfaceTypeEmbeddings) {
 			return body, nil
 		}
-		return e.convertEmbeddingResponseBody(clientAdapter, providerAdapter, body)
+		return e.convertEmbeddingResponseBody(clientAdapter, providerAdapter, body, modelOverride)
 	case InterfaceTypeRerank:
 		if !clientAdapter.SupportsInterface(InterfaceTypeRerank) || !providerAdapter.SupportsInterface(InterfaceTypeRerank) {
 			return body, nil
 		}
-		return e.convertRerankResponseBody(clientAdapter, providerAdapter, body)
+		return e.convertRerankResponseBody(clientAdapter, providerAdapter, body, modelOverride)
 	default:
 		return body, nil
 	}
@@ -241,11 +289,14 @@ func (e *ConversionEngine) convertChatBody(clientAdapter, providerAdapter Protoc
 	return encoded, nil
 }
 
-func (e *ConversionEngine) convertChatResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte) ([]byte, error) {
+func (e *ConversionEngine) convertChatResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte, modelOverride string) ([]byte, error) {
 	canonicalResp, err := providerAdapter.DecodeResponse(body)
 	if err != nil {
 		return nil, NewConversionError(ErrorCodeJSONParseError, "解码响应失败").WithCause(err)
 	}
+	if modelOverride != "" {
+		canonicalResp.Model = modelOverride
+	}
 	encoded, err := clientAdapter.EncodeResponse(canonicalResp)
 	if err != nil {
 		return nil, NewConversionError(ErrorCodeEncodingFailure, "编码响应失败").WithCause(err)
@@ -290,12 +341,15 @@ func (e *ConversionEngine) convertEmbeddingBody(clientAdapter, providerAdapter P
 	return providerAdapter.EncodeEmbeddingRequest(req, provider)
 }
 
-func (e *ConversionEngine) convertEmbeddingResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte) ([]byte, error) {
+func (e *ConversionEngine) convertEmbeddingResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte, modelOverride string) ([]byte, error) {
 	resp, err := providerAdapter.DecodeEmbeddingResponse(body)
 	if err != nil {
 		e.logger.Warn("解码 Embedding 响应失败，返回原始响应", zap.String("error", err.Error()))
 		return body, nil
 	}
+	if modelOverride != "" {
+		resp.Model = modelOverride
+	}
 	return clientAdapter.EncodeEmbeddingResponse(resp)
 }
 
@@ -308,11 +362,14 @@ func (e *ConversionEngine) convertRerankBody(clientAdapter, providerAdapter Prot
 	return providerAdapter.EncodeRerankRequest(req, provider)
 }
 
-func (e *ConversionEngine) convertRerankResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte) ([]byte, error) {
+func (e *ConversionEngine) convertRerankResponseBody(clientAdapter, providerAdapter ProtocolAdapter, body []byte, modelOverride string) ([]byte, error) {
 	resp, err := providerAdapter.DecodeRerankResponse(body)
 	if err != nil {
 		return body, nil
 	}
+	if modelOverride != "" {
+		resp.Model = modelOverride
+	}
 	return clientAdapter.EncodeRerankResponse(resp)
 }
 

backend/internal/conversion/engine_supplemental_test.go

@@ -113,7 +113,7 @@ func TestConvertHttpResponse_CrossProtocol(t *testing.T) {
 
 	result, err := engine.ConvertHttpResponse(HTTPResponseSpec{
 		StatusCode: 200, Body: []byte(`{"id":"resp-1"}`),
-	}, "client", "provider", InterfaceTypeChat)
+	}, "client", "provider", InterfaceTypeChat, "")
 	require.NoError(t, err)
 	assert.Equal(t, 200, result.StatusCode)
 	assert.Contains(t, string(result.Body), "resp-1")
@@ -129,7 +129,7 @@ func TestConvertHttpResponse_DecodeError(t *testing.T) {
 	_ = engine.RegisterAdapter(providerAdapter)
 	_ = engine.RegisterAdapter(newMockAdapter("client", false))
 
-	_, err := engine.ConvertHttpResponse(HTTPResponseSpec{Body: []byte(`{}`)}, "client", "provider", InterfaceTypeChat)
+	_, err := engine.ConvertHttpResponse(HTTPResponseSpec{Body: []byte(`{}`)}, "client", "provider", InterfaceTypeChat, "")
 	assert.Error(t, err)
 }
 
@@ -189,7 +189,7 @@ func TestConvertHttpResponse_EmbeddingInterface(t *testing.T) {
 
 	result, err := engine.ConvertHttpResponse(HTTPResponseSpec{
 		StatusCode: 200, Body: []byte(`{"object":"list","data":[],"model":"test"}`),
-	}, "client", "provider", InterfaceTypeEmbeddings)
+	}, "client", "provider", InterfaceTypeEmbeddings, "")
 	require.NoError(t, err)
 	assert.NotNil(t, result)
 }
@@ -207,7 +207,7 @@ func TestConvertHttpResponse_RerankInterface(t *testing.T) {
 
 	result, err := engine.ConvertHttpResponse(HTTPResponseSpec{
 		StatusCode: 200, Body: []byte(`{"results":[],"model":"test"}`),
-	}, "client", "provider", InterfaceTypeRerank)
+	}, "client", "provider", InterfaceTypeRerank, "")
 	require.NoError(t, err)
 	assert.NotNil(t, result)
 }
@@ -242,7 +242,7 @@ func TestConvertHttpResponse_ModelsInterface(t *testing.T) {
 
 	result, err := engine.ConvertHttpResponse(HTTPResponseSpec{
 		StatusCode: 200, Body: []byte(`{"object":"list","data":[]}`),
-	}, "client", "provider", InterfaceTypeModels)
+	}, "client", "provider", InterfaceTypeModels, "")
 	require.NoError(t, err)
 	assert.NotNil(t, result)
 }
@@ -259,7 +259,7 @@ func TestConvertHttpResponse_ModelInfoInterface(t *testing.T) {
 
 	result, err := engine.ConvertHttpResponse(HTTPResponseSpec{
 		StatusCode: 200, Body: []byte(`{"id":"gpt-4","object":"model"}`),
-	}, "client", "provider", InterfaceTypeModelInfo)
+	}, "client", "provider", InterfaceTypeModelInfo, "")
 	require.NoError(t, err)
 	assert.NotNil(t, result)
 }

backend/internal/conversion/engine_test.go

@@ -13,16 +13,18 @@ import (
 
 // mockProtocolAdapter 模拟协议适配器
 type mockProtocolAdapter struct {
-	protocolName    string
+	protocolName     string
-	passthrough     bool
+	passthrough      bool
-	ifaceType       InterfaceType
+	ifaceType        InterfaceType
-	supportsIface   map[InterfaceType]bool
+	supportsIface    map[InterfaceType]bool
-	decodeReqFn     func([]byte) (*canonical.CanonicalRequest, error)
+	decodeReqFn      func([]byte) (*canonical.CanonicalRequest, error)
-	encodeReqFn     func(*canonical.CanonicalRequest, *TargetProvider) ([]byte, error)
+	encodeReqFn      func(*canonical.CanonicalRequest, *TargetProvider) ([]byte, error)
-	decodeRespFn    func([]byte) (*canonical.CanonicalResponse, error)
+	decodeRespFn     func([]byte) (*canonical.CanonicalResponse, error)
-	encodeRespFn    func(*canonical.CanonicalResponse) ([]byte, error)
+	encodeRespFn     func(*canonical.CanonicalResponse) ([]byte, error)
-	streamDecoderFn func() StreamDecoder
+	streamDecoderFn  func() StreamDecoder
-	streamEncoderFn func() StreamEncoder
+	streamEncoderFn  func() StreamEncoder
+	rewriteReqFn     func([]byte, string, InterfaceType) ([]byte, error)
+	rewriteRespFn    func([]byte, string, InterfaceType) ([]byte, error)
 }
 
 func newMockAdapter(name string, passthrough bool) *mockProtocolAdapter {
@@ -155,6 +157,28 @@ func (m *mockProtocolAdapter) EncodeRerankResponse(resp *canonical.CanonicalRera
 	return json.Marshal(resp)
 }
 
+func (m *mockProtocolAdapter) ExtractUnifiedModelID(nativePath string) (string, error) {
+	return "", nil
+}
+
+func (m *mockProtocolAdapter) ExtractModelName(body []byte, ifaceType InterfaceType) (string, error) {
+	return "", nil
+}
+
+func (m *mockProtocolAdapter) RewriteRequestModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error) {
+	if m.rewriteReqFn != nil {
+		return m.rewriteReqFn(body, newModel, ifaceType)
+	}
+	return body, nil
+}
+
+func (m *mockProtocolAdapter) RewriteResponseModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error) {
+	if m.rewriteRespFn != nil {
+		return m.rewriteRespFn(body, newModel, ifaceType)
+	}
+	return body, nil
+}
+
 // noopStreamDecoder 空流式解码器
 type noopStreamDecoder struct{}
 
@@ -309,7 +333,7 @@ func TestConvertHttpResponse_Passthrough(t *testing.T) {
 		Body:       []byte(`{"id":"123"}`),
 	}
 
-	result, err := engine.ConvertHttpResponse(spec, "openai", "openai", InterfaceTypeChat)
+	result, err := engine.ConvertHttpResponse(spec, "openai", "openai", InterfaceTypeChat, "")
 	require.NoError(t, err)
 	assert.Equal(t, 200, result.StatusCode)
 	assert.Equal(t, spec.Body, result.Body)
@@ -320,7 +344,7 @@ func TestCreateStreamConverter_Passthrough(t *testing.T) {
 	engine := NewConversionEngine(registry, nil)
 	_ = engine.RegisterAdapter(newMockAdapter("openai", true))
 
-	converter, err := engine.CreateStreamConverter("openai", "openai")
+	converter, err := engine.CreateStreamConverter("openai", "openai", "", InterfaceTypeChat)
 	require.NoError(t, err)
 	_, ok := converter.(*PassthroughStreamConverter)
 	assert.True(t, ok)
@@ -332,7 +356,7 @@ func TestCreateStreamConverter_Canonical(t *testing.T) {
 	_ = engine.RegisterAdapter(newMockAdapter("client", false))
 	_ = engine.RegisterAdapter(newMockAdapter("provider", false))
 
-	converter, err := engine.CreateStreamConverter("client", "provider")
+	converter, err := engine.CreateStreamConverter("client", "provider", "", InterfaceTypeChat)
 	require.NoError(t, err)
 	_, ok := converter.(*CanonicalStreamConverter)
 	assert.True(t, ok)
@@ -380,3 +404,230 @@ func TestRegistry_GetNonExistent(t *testing.T) {
 	assert.Error(t, err)
 	assert.Contains(t, err.Error(), "未找到适配器")
 }
+
+// ============ modelOverride 测试 ============
+
+func TestConvertHttpResponse_ModelOverride_CrossProtocol(t *testing.T) {
+	registry := NewMemoryRegistry()
+	engine := NewConversionEngine(registry, nil)
+
+	clientAdapter := newMockAdapter("client", false)
+	clientAdapter.encodeRespFn = func(resp *canonical.CanonicalResponse) ([]byte, error) {
+		return json.Marshal(map[string]any{"model": resp.Model})
+	}
+	_ = engine.RegisterAdapter(clientAdapter)
+
+	providerAdapter := newMockAdapter("provider", false)
+	providerAdapter.decodeRespFn = func(raw []byte) (*canonical.CanonicalResponse, error) {
+		return &canonical.CanonicalResponse{ID: "test", Model: "native-model", Content: []canonical.ContentBlock{canonical.NewTextBlock("hi")}}, nil
+	}
+	_ = engine.RegisterAdapter(providerAdapter)
+
+	spec := HTTPResponseSpec{
+		StatusCode: 200,
+		Body:       []byte(`{"model":"native-model"}`),
+	}
+
+	result, err := engine.ConvertHttpResponse(spec, "client", "provider", InterfaceTypeChat, "provider/gpt-4")
+	require.NoError(t, err)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(result.Body, &resp))
+	assert.Equal(t, "provider/gpt-4", resp["model"])
+}
+
+func TestConvertHttpResponse_ModelOverride_SameProtocol(t *testing.T) {
+	registry := NewMemoryRegistry()
+	engine := NewConversionEngine(registry, nil)
+
+	// 使用真实 OpenAI adapter 验证 Smart Passthrough 改写
+	openaiAdapter := newMockAdapter("openai", true)
+	openaiAdapter.rewriteRespFn = func(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error) {
+		var m map[string]json.RawMessage
+		if err := json.Unmarshal(body, &m); err != nil {
+			return nil, err
+		}
+		m["model"], _ = json.Marshal(newModel)
+		return json.Marshal(m)
+	}
+	_ = engine.RegisterAdapter(openaiAdapter)
+
+	spec := HTTPResponseSpec{
+		StatusCode: 200,
+		Body:       []byte(`{"id":"resp-1","model":"gpt-4"}`),
+	}
+
+	result, err := engine.ConvertHttpResponse(spec, "openai", "openai", InterfaceTypeChat, "openai/gpt-4")
+	require.NoError(t, err)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(result.Body, &resp))
+	assert.Equal(t, "openai/gpt-4", resp["model"])
+	assert.Equal(t, "resp-1", resp["id"])
+}
+
+func TestCreateStreamConverter_ModelOverride_SmartPassthrough(t *testing.T) {
+	registry := NewMemoryRegistry()
+	engine := NewConversionEngine(registry, nil)
+
+	openaiAdapter := newMockAdapter("openai", true)
+	openaiAdapter.rewriteRespFn = func(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error) {
+		var m map[string]json.RawMessage
+		if err := json.Unmarshal(body, &m); err != nil {
+			return nil, err
+		}
+		m["model"], _ = json.Marshal(newModel)
+		return json.Marshal(m)
+	}
+	_ = engine.RegisterAdapter(openaiAdapter)
+
+	converter, err := engine.CreateStreamConverter("openai", "openai", "openai/gpt-4", InterfaceTypeChat)
+	require.NoError(t, err)
+
+	_, ok := converter.(*SmartPassthroughStreamConverter)
+	assert.True(t, ok)
+
+	// 验证 chunk 改写
+	chunks := converter.ProcessChunk([]byte(`{"model":"gpt-4","choices":[]}`))
+	require.Len(t, chunks, 1)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(chunks[0], &resp))
+	assert.Equal(t, "openai/gpt-4", resp["model"])
+}
+
+func TestCreateStreamConverter_ModelOverride_CrossProtocol(t *testing.T) {
+	registry := NewMemoryRegistry()
+	engine := NewConversionEngine(registry, nil)
+
+	// provider adapter 解码出含 model 的流式事件
+	providerAdapter := newMockAdapter("provider", false)
+	providerAdapter.streamDecoderFn = func() StreamDecoder {
+		return &engineTestStreamDecoder{
+			processFn: func(raw []byte) []canonical.CanonicalStreamEvent {
+				return []canonical.CanonicalStreamEvent{
+					canonical.NewMessageStartEvent("msg-1", "native-model"),
+					canonical.NewContentBlockStartEvent(0, canonical.StreamContentBlock{Type: "text", Text: "hi"}),
+					canonical.NewMessageStopEvent(),
+				}
+			},
+		}
+	}
+	_ = engine.RegisterAdapter(providerAdapter)
+
+	// client adapter 编码时输出 model 字段
+	clientAdapter := newMockAdapter("client", false)
+	clientAdapter.streamEncoderFn = func() StreamEncoder {
+		return &engineTestStreamEncoder{
+			encodeFn: func(event canonical.CanonicalStreamEvent) [][]byte {
+				if event.Message != nil {
+					data, _ := json.Marshal(map[string]string{
+						"type":  string(event.Type),
+						"model": event.Message.Model,
+					})
+					return [][]byte{data}
+				}
+				data, _ := json.Marshal(map[string]string{"type": string(event.Type)})
+				return [][]byte{data}
+			},
+		}
+	}
+	_ = engine.RegisterAdapter(clientAdapter)
+
+	converter, err := engine.CreateStreamConverter("client", "provider", "provider/gpt-4", InterfaceTypeChat)
+	require.NoError(t, err)
+
+	// 验证类型是 CanonicalStreamConverter
+	_, ok := converter.(*CanonicalStreamConverter)
+	assert.True(t, ok)
+
+	// 处理一个 chunk，验证 model 被覆写为统一模型 ID
+	chunks := converter.ProcessChunk([]byte("raw"))
+	require.Len(t, chunks, 3) // message_start + content_block_start + message_stop
+
+	var startEvent map[string]string
+	require.NoError(t, json.Unmarshal(chunks[0], &startEvent))
+	assert.Equal(t, "provider/gpt-4", startEvent["model"], "跨协议流式中 modelOverride 应覆写 Message.Model")
+}
+
+func TestCreateStreamConverter_ModelOverride_CrossProtocol_Empty(t *testing.T) {
+	registry := NewMemoryRegistry()
+	engine := NewConversionEngine(registry, nil)
+
+	providerAdapter := newMockAdapter("provider", false)
+	providerAdapter.streamDecoderFn = func() StreamDecoder {
+		return &engineTestStreamDecoder{
+			processFn: func(raw []byte) []canonical.CanonicalStreamEvent {
+				return []canonical.CanonicalStreamEvent{
+					canonical.NewMessageStartEvent("msg-1", "native-model"),
+				}
+			},
+		}
+	}
+	_ = engine.RegisterAdapter(providerAdapter)
+
+	clientAdapter := newMockAdapter("client", false)
+	clientAdapter.streamEncoderFn = func() StreamEncoder {
+		return &engineTestStreamEncoder{
+			encodeFn: func(event canonical.CanonicalStreamEvent) [][]byte {
+				if event.Message != nil {
+					data, _ := json.Marshal(map[string]string{
+						"model": event.Message.Model,
+					})
+					return [][]byte{data}
+				}
+				return nil
+			},
+		}
+	}
+	_ = engine.RegisterAdapter(clientAdapter)
+
+	// modelOverride 为空，不应覆写
+	converter, err := engine.CreateStreamConverter("client", "provider", "", InterfaceTypeChat)
+	require.NoError(t, err)
+
+	chunks := converter.ProcessChunk([]byte("raw"))
+	require.Len(t, chunks, 1)
+
+	var resp map[string]string
+	require.NoError(t, json.Unmarshal(chunks[0], &resp))
+	assert.Equal(t, "native-model", resp["model"], "modelOverride 为空时不应覆写")
+}
+
+// engineTestStreamDecoder 可控的流式解码器（用于 engine_test）
+type engineTestStreamDecoder struct {
+	processFn func([]byte) []canonical.CanonicalStreamEvent
+	flushFn   func() []canonical.CanonicalStreamEvent
+}
+
+func (d *engineTestStreamDecoder) ProcessChunk(raw []byte) []canonical.CanonicalStreamEvent {
+	if d.processFn != nil {
+		return d.processFn(raw)
+	}
+	return nil
+}
+func (d *engineTestStreamDecoder) Flush() []canonical.CanonicalStreamEvent {
+	if d.flushFn != nil {
+		return d.flushFn()
+	}
+	return nil
+}
+
+// engineTestStreamEncoder 可控的流式编码器（用于 engine_test）
+type engineTestStreamEncoder struct {
+	encodeFn func(canonical.CanonicalStreamEvent) [][]byte
+	flushFn  func() [][]byte
+}
+
+func (e *engineTestStreamEncoder) EncodeEvent(event canonical.CanonicalStreamEvent) [][]byte {
+	if e.encodeFn != nil {
+		return e.encodeFn(event)
+	}
+	return nil
+}
+func (e *engineTestStreamEncoder) Flush() [][]byte {
+	if e.flushFn != nil {
+		return e.flushFn()
+	}
+	return nil
+}

backend/internal/conversion/openai/adapter.go

@@ -2,6 +2,7 @@ package openai
 
 import (
 	"encoding/json"
+	"fmt"
 	"strings"
 
 	"nex/backend/internal/conversion"
@@ -43,13 +44,13 @@ func (a *Adapter) DetectInterfaceType(nativePath string) conversion.InterfaceTyp
 	}
 }
 
-// isModelInfoPath 判断是否为模型详情路径（/v1/models/{id}）
+// isModelInfoPath 判断是否为模型详情路径（/v1/models/{id}，允许 id 含 /）
 func isModelInfoPath(path string) bool {
 	if !strings.HasPrefix(path, "/v1/models/") {
 		return false
 	}
 	suffix := path[len("/v1/models/"):]
-	return suffix != "" && !strings.Contains(suffix, "/")
+	return suffix != ""
 }
 
 // BuildUrl 根据接口类型构建 URL
@@ -216,3 +217,80 @@ func (a *Adapter) DecodeRerankResponse(raw []byte) (*canonical.CanonicalRerankRe
 func (a *Adapter) EncodeRerankResponse(resp *canonical.CanonicalRerankResponse) ([]byte, error) {
 	return encodeRerankResponse(resp)
 }
+
+// ExtractUnifiedModelID 从路径中提取统一模型 ID（/v1/models/{provider_id}/{model_name}）
+func (a *Adapter) ExtractUnifiedModelID(nativePath string) (string, error) {
+	if !strings.HasPrefix(nativePath, "/v1/models/") {
+		return "", fmt.Errorf("不是模型详情路径: %s", nativePath)
+	}
+	suffix := nativePath[len("/v1/models/"):]
+	if suffix == "" {
+		return "", fmt.Errorf("路径缺少模型 ID")
+	}
+	return suffix, nil
+}
+
+// locateModelFieldInRequest 定位请求体中 model 字段的值并提供改写函数
+func locateModelFieldInRequest(body []byte, ifaceType conversion.InterfaceType) (string, func(string) ([]byte, error), error) {
+	var m map[string]json.RawMessage
+	if err := json.Unmarshal(body, &m); err != nil {
+		return "", nil, err
+	}
+
+	switch ifaceType {
+	case conversion.InterfaceTypeChat, conversion.InterfaceTypeEmbeddings, conversion.InterfaceTypeRerank:
+		raw, exists := m["model"]
+		if !exists {
+			return "", nil, fmt.Errorf("请求体中缺少 model 字段")
+		}
+		var current string
+		if err := json.Unmarshal(raw, &current); err != nil {
+			return "", nil, err
+		}
+		rewriteFunc := func(newModel string) ([]byte, error) {
+			m["model"], _ = json.Marshal(newModel)
+			return json.Marshal(m)
+		}
+		return current, rewriteFunc, nil
+	default:
+		return "", nil, fmt.Errorf("不支持的接口类型: %s", ifaceType)
+	}
+}
+
+// ExtractModelName 从请求体中提取 model 值
+func (a *Adapter) ExtractModelName(body []byte, ifaceType conversion.InterfaceType) (string, error) {
+	model, _, err := locateModelFieldInRequest(body, ifaceType)
+	return model, err
+}
+
+// RewriteRequestModelName 最小化改写请求体中的 model 字段
+func (a *Adapter) RewriteRequestModelName(body []byte, newModel string, ifaceType conversion.InterfaceType) ([]byte, error) {
+	_, rewriteFunc, err := locateModelFieldInRequest(body, ifaceType)
+	if err != nil {
+		return nil, err
+	}
+	return rewriteFunc(newModel)
+}
+
+// RewriteResponseModelName 最小化改写响应体中的 model 字段
+func (a *Adapter) RewriteResponseModelName(body []byte, newModel string, ifaceType conversion.InterfaceType) ([]byte, error) {
+	var m map[string]json.RawMessage
+	if err := json.Unmarshal(body, &m); err != nil {
+		return nil, err
+	}
+
+	switch ifaceType {
+	case conversion.InterfaceTypeChat, conversion.InterfaceTypeEmbeddings:
+		// Chat/Embedding 响应必须有 model 字段（协议要求），存在则改写，不存在则添加
+		m["model"], _ = json.Marshal(newModel)
+		return json.Marshal(m)
+	case conversion.InterfaceTypeRerank:
+		// Rerank 响应：存在 model 字段则改写，不存在则不添加
+		if _, exists := m["model"]; exists {
+			m["model"], _ = json.Marshal(newModel)
+		}
+		return json.Marshal(m)
+	default:
+		return body, nil
+	}
+}

backend/internal/conversion/openai/adapter_test.go

@@ -121,7 +121,7 @@ func TestIsModelInfoPath(t *testing.T) {
 		{"model_info", "/v1/models/gpt-4", true},
 		{"model_info_with_dots", "/v1/models/gpt-4.1-preview", true},
 		{"models_list", "/v1/models", false},
-		{"nested_path", "/v1/models/gpt-4/versions", false},
+		{"nested_path", "/v1/models/gpt-4/versions", true},
 		{"empty_suffix", "/v1/models/", false},
 		{"unrelated", "/v1/chat/completions", false},
 		{"partial_prefix", "/v1/model", false},

backend/internal/conversion/openai/adapter_unified_test.go

@@ -0,0 +1,360 @@
+package openai
+
+import (
+	"encoding/json"
+	"testing"
+
+	"nex/backend/internal/conversion"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// ---------------------------------------------------------------------------
+// ExtractUnifiedModelID
+// ---------------------------------------------------------------------------
+
+func TestExtractUnifiedModelID(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("standard_path", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/openai/gpt-4")
+		require.NoError(t, err)
+		assert.Equal(t, "openai/gpt-4", id)
+	})
+
+	t.Run("multi_segment_path", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/azure/accounts/org/models/gpt-4")
+		require.NoError(t, err)
+		assert.Equal(t, "azure/accounts/org/models/gpt-4", id)
+	})
+
+	t.Run("single_segment", func(t *testing.T) {
+		id, err := a.ExtractUnifiedModelID("/v1/models/gpt-4")
+		require.NoError(t, err)
+		assert.Equal(t, "gpt-4", id)
+	})
+
+	t.Run("non_model_path", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/chat/completions")
+		require.Error(t, err)
+	})
+
+	t.Run("empty_suffix", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/models/")
+		require.Error(t, err)
+	})
+
+	t.Run("models_list_no_slash", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/models")
+		require.Error(t, err)
+	})
+
+	t.Run("unrelated_path", func(t *testing.T) {
+		_, err := a.ExtractUnifiedModelID("/v1/other")
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// ExtractModelName
+// ---------------------------------------------------------------------------
+
+func TestExtractModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat", func(t *testing.T) {
+		body := []byte(`{"model":"openai/gpt-4","messages":[]}`)
+		model, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/gpt-4", model)
+	})
+
+	t.Run("embedding", func(t *testing.T) {
+		body := []byte(`{"model":"openai/text-embedding","input":"hello"}`)
+		model, err := a.ExtractModelName(body, conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/text-embedding", model)
+	})
+
+	t.Run("rerank", func(t *testing.T) {
+		body := []byte(`{"model":"openai/rerank","query":"test"}`)
+		model, err := a.ExtractModelName(body, conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/rerank", model)
+	})
+
+	t.Run("no_model_field", func(t *testing.T) {
+		body := []byte(`{"messages":[]}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("unsupported_interface_type", func(t *testing.T) {
+		body := []byte(`{"model":"openai/gpt-4"}`)
+		_, err := a.ExtractModelName(body, conversion.InterfaceTypePassthrough)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// RewriteRequestModelName
+// ---------------------------------------------------------------------------
+
+func TestRewriteRequestModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat", func(t *testing.T) {
+		body := []byte(`{"model":"openai/gpt-4","messages":[]}`)
+		rewritten, err := a.RewriteRequestModelName(body, "gpt-4", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "gpt-4", m["model"])
+
+		// messages field preserved
+		msgs, ok := m["messages"]
+		require.True(t, ok)
+		msgsArr, ok := msgs.([]interface{})
+		require.True(t, ok)
+		assert.Len(t, msgsArr, 0)
+	})
+
+	t.Run("preserves_unknown_fields", func(t *testing.T) {
+		body := []byte(`{"model":"openai/gpt-4","temperature":0.7}`)
+		rewritten, err := a.RewriteRequestModelName(body, "gpt-4", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "gpt-4", m["model"])
+		assert.Equal(t, 0.7, m["temperature"])
+	})
+
+	t.Run("embedding", func(t *testing.T) {
+		body := []byte(`{"model":"openai/text-embedding","input":"hello"}`)
+		rewritten, err := a.RewriteRequestModelName(body, "text-embedding", conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "text-embedding", m["model"])
+		assert.Equal(t, "hello", m["input"])
+	})
+
+	t.Run("rerank", func(t *testing.T) {
+		body := []byte(`{"model":"openai/rerank","query":"test"}`)
+		rewritten, err := a.RewriteRequestModelName(body, "rerank", conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "rerank", m["model"])
+		assert.Equal(t, "test", m["query"])
+	})
+
+	t.Run("no_model_field", func(t *testing.T) {
+		body := []byte(`{"messages":[]}`)
+		_, err := a.RewriteRequestModelName(body, "gpt-4", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.RewriteRequestModelName(body, "gpt-4", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+
+	t.Run("unsupported_interface_type", func(t *testing.T) {
+		body := []byte(`{"model":"openai/gpt-4"}`)
+		_, err := a.RewriteRequestModelName(body, "gpt-4", conversion.InterfaceTypePassthrough)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// RewriteResponseModelName
+// ---------------------------------------------------------------------------
+
+func TestRewriteResponseModelName(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat_existing_model", func(t *testing.T) {
+		body := []byte(`{"model":"gpt-4","choices":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/gpt-4", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "openai/gpt-4", m["model"])
+
+		choices, ok := m["choices"]
+		require.True(t, ok)
+		choicesArr, ok := choices.([]interface{})
+		require.True(t, ok)
+		assert.Len(t, choicesArr, 0)
+	})
+
+	t.Run("chat_without_model_field", func(t *testing.T) {
+		body := []byte(`{"choices":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/gpt-4", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "openai/gpt-4", m["model"])
+
+		choices, ok := m["choices"]
+		require.True(t, ok)
+		choicesArr, ok := choices.([]interface{})
+		require.True(t, ok)
+		assert.Len(t, choicesArr, 0)
+	})
+
+	t.Run("rerank_existing_model", func(t *testing.T) {
+		body := []byte(`{"model":"rerank","results":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/rerank", conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "openai/rerank", m["model"])
+	})
+
+	t.Run("rerank_without_model_field_should_not_add", func(t *testing.T) {
+		body := []byte(`{"results":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/rerank", conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		_, hasModel := m["model"]
+		assert.False(t, hasModel, "rerank response without model field should not have one added")
+	})
+
+	t.Run("embedding_existing_model", func(t *testing.T) {
+		body := []byte(`{"model":"text-embedding","data":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/text-embedding", conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "openai/text-embedding", m["model"])
+	})
+
+	t.Run("embedding_without_model_field_adds", func(t *testing.T) {
+		body := []byte(`{"data":[]}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/text-embedding", conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, "openai/text-embedding", m["model"])
+	})
+
+	t.Run("passthrough_returns_body_unchanged", func(t *testing.T) {
+		body := []byte(`{"model":"gpt-4"}`)
+		rewritten, err := a.RewriteResponseModelName(body, "openai/gpt-4", conversion.InterfaceTypePassthrough)
+		require.NoError(t, err)
+		assert.Equal(t, string(body), string(rewritten))
+	})
+
+	t.Run("invalid_json", func(t *testing.T) {
+		body := []byte(`{invalid}`)
+		_, err := a.RewriteResponseModelName(body, "openai/gpt-4", conversion.InterfaceTypeChat)
+		require.Error(t, err)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// ExtractModelName and RewriteRequest consistency
+// ---------------------------------------------------------------------------
+
+func TestExtractModelNameAndRewriteRequestConsistency(t *testing.T) {
+	a := NewAdapter()
+
+	t.Run("chat_round_trip", func(t *testing.T) {
+		original := []byte(`{"model":"openai/gpt-4","messages":[],"temperature":0.7}`)
+
+		// Extract the unified model ID from the body
+		extracted, err := a.ExtractModelName(original, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/gpt-4", extracted)
+
+		// Rewrite to the native model name
+		rewritten, err := a.RewriteRequestModelName(original, "gpt-4", conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+
+		// Extract again from the rewritten body to verify the same location was targeted
+		afterRewrite, err := a.ExtractModelName(rewritten, conversion.InterfaceTypeChat)
+		require.NoError(t, err)
+		assert.Equal(t, "gpt-4", afterRewrite)
+
+		// Verify other fields are preserved
+		var m map[string]interface{}
+		require.NoError(t, json.Unmarshal(rewritten, &m))
+		assert.Equal(t, 0.7, m["temperature"])
+	})
+
+	t.Run("embedding_round_trip", func(t *testing.T) {
+		original := []byte(`{"model":"openai/text-embedding","input":"hello"}`)
+
+		extracted, err := a.ExtractModelName(original, conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/text-embedding", extracted)
+
+		rewritten, err := a.RewriteRequestModelName(original, "text-embedding", conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+
+		afterRewrite, err := a.ExtractModelName(rewritten, conversion.InterfaceTypeEmbeddings)
+		require.NoError(t, err)
+		assert.Equal(t, "text-embedding", afterRewrite)
+	})
+
+	t.Run("rerank_round_trip", func(t *testing.T) {
+		original := []byte(`{"model":"openai/rerank","query":"test"}`)
+
+		extracted, err := a.ExtractModelName(original, conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+		assert.Equal(t, "openai/rerank", extracted)
+
+		rewritten, err := a.RewriteRequestModelName(original, "rerank", conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+
+		afterRewrite, err := a.ExtractModelName(rewritten, conversion.InterfaceTypeRerank)
+		require.NoError(t, err)
+		assert.Equal(t, "rerank", afterRewrite)
+	})
+}
+
+// ---------------------------------------------------------------------------
+// isModelInfoPath (additional unified model ID cases)
+// ---------------------------------------------------------------------------
+
+func TestIsModelInfoPath_UnifiedModelID(t *testing.T) {
+	tests := []struct {
+		name     string
+		path     string
+		expected bool
+	}{
+		{"simple_model_id", "/v1/models/gpt-4", true},
+		{"unified_model_id_with_slash", "/v1/models/openai/gpt-4", true},
+		{"models_list", "/v1/models", false},
+		{"models_list_trailing_slash", "/v1/models/", false},
+		{"chat_completions", "/v1/chat/completions", false},
+		{"deeply_nested", "/v1/models/azure/eastus/deployments/my-dept/models/gpt-4", true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			assert.Equal(t, tt.expected, isModelInfoPath(tt.path))
+		})
+	}
+}

backend/internal/conversion/stream.go

@@ -38,14 +38,52 @@ func (c *PassthroughStreamConverter) Flush() [][]byte {
 	return nil
 }
 
+// SmartPassthroughStreamConverter 同协议 Smart Passthrough 流式转换器
+// 逐 chunk 改写 model 字段
+type SmartPassthroughStreamConverter struct {
+	adapter       ProtocolAdapter
+	modelOverride string
+	interfaceType InterfaceType
+}
+
+// NewSmartPassthroughStreamConverter 创建 Smart Passthrough 流式转换器
+func NewSmartPassthroughStreamConverter(adapter ProtocolAdapter, modelOverride string, interfaceType InterfaceType) *SmartPassthroughStreamConverter {
+	return &SmartPassthroughStreamConverter{
+		adapter:       adapter,
+		modelOverride: modelOverride,
+		interfaceType: interfaceType,
+	}
+}
+
+// ProcessChunk 改写 chunk 中的 model 字段
+func (c *SmartPassthroughStreamConverter) ProcessChunk(rawChunk []byte) [][]byte {
+	if len(rawChunk) == 0 {
+		return nil
+	}
+
+	rewrittenChunk, err := c.adapter.RewriteResponseModelName(rawChunk, c.modelOverride, c.interfaceType)
+	if err != nil {
+		// 改写失败，返回原始 chunk
+		return [][]byte{rawChunk}
+	}
+
+	return [][]byte{rewrittenChunk}
+}
+
+// Flush 无缓冲数据
+func (c *SmartPassthroughStreamConverter) Flush() [][]byte {
+	return nil
+}
+
 // CanonicalStreamConverter 跨协议规范流式转换器
 type CanonicalStreamConverter struct {
-	decoder  StreamDecoder
+	decoder          StreamDecoder
-	encoder  StreamEncoder
+	encoder          StreamEncoder
-	chain    *MiddlewareChain
+	chain            *MiddlewareChain
-	ctx      ConversionContext
+	ctx              ConversionContext
-	clientProtocol  string
+	clientProtocol   string
 	providerProtocol string
+	modelOverride    string
 }
 
 // NewCanonicalStreamConverter 创建规范流式转换器
@@ -57,18 +95,19 @@ func NewCanonicalStreamConverter(decoder StreamDecoder, encoder StreamEncoder) *
 }
 
 // NewCanonicalStreamConverterWithMiddleware 创建带中间件的规范流式转换器
-func NewCanonicalStreamConverterWithMiddleware(decoder StreamDecoder, encoder StreamEncoder, chain *MiddlewareChain, ctx ConversionContext, clientProtocol, providerProtocol string) *CanonicalStreamConverter {
+func NewCanonicalStreamConverterWithMiddleware(decoder StreamDecoder, encoder StreamEncoder, chain *MiddlewareChain, ctx ConversionContext, clientProtocol, providerProtocol, modelOverride string) *CanonicalStreamConverter {
 	return &CanonicalStreamConverter{
-		decoder:         decoder,
+		decoder:          decoder,
-		encoder:         encoder,
+		encoder:          encoder,
-		chain:           chain,
+		chain:            chain,
-		ctx:             ctx,
+		ctx:              ctx,
-		clientProtocol:  clientProtocol,
+		clientProtocol:   clientProtocol,
 		providerProtocol: providerProtocol,
+		modelOverride:    modelOverride,
 	}
 }
 
-// ProcessChunk 解码 → 中间件 → 编码管道
+// ProcessChunk 解码 → 中间件 → modelOverride → 编码管道
 func (c *CanonicalStreamConverter) ProcessChunk(rawChunk []byte) [][]byte {
 	events := c.decoder.ProcessChunk(rawChunk)
 	var result [][]byte
@@ -80,6 +119,7 @@ func (c *CanonicalStreamConverter) ProcessChunk(rawChunk []byte) [][]byte {
 			}
 			events[i] = *processed
 		}
+		c.applyModelOverride(&events[i])
 		chunks := c.encoder.EncodeEvent(events[i])
 		result = append(result, chunks...)
 	}
@@ -98,6 +138,7 @@ func (c *CanonicalStreamConverter) Flush() [][]byte {
 			}
 			events[i] = *processed
 		}
+		c.applyModelOverride(&events[i])
 		chunks := c.encoder.EncodeEvent(events[i])
 		result = append(result, chunks...)
 	}
@@ -105,3 +146,10 @@ func (c *CanonicalStreamConverter) Flush() [][]byte {
 	result = append(result, encoderChunks...)
 	return result
 }
+
+// applyModelOverride 在跨协议场景下覆写流式事件中的 Model 字段
+func (c *CanonicalStreamConverter) applyModelOverride(event *canonical.CanonicalStreamEvent) {
+	if c.modelOverride != "" && event.Message != nil {
+		event.Message.Model = c.modelOverride
+	}
+}

backend/internal/conversion/stream_test.go

@@ -93,7 +93,7 @@ func TestCanonicalStreamConverter_WithMiddleware(t *testing.T) {
 	chain.Use(&recordingMiddleware{name: "mw1", records: &records})
 	ctx := NewConversionContext(InterfaceTypeChat)
 
-	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic")
+	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic", "")
 	result := converter.ProcessChunk([]byte("raw"))
 
 	assert.Len(t, result, 1)
@@ -143,7 +143,7 @@ func TestCanonicalStreamConverter_MiddlewareError_Continue(t *testing.T) {
 	chain.Use(&errorMiddleware{})
 	ctx := NewConversionContext(InterfaceTypeChat)
 
-	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic")
+	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic", "")
 	result := converter.ProcessChunk([]byte("raw"))
 
 	assert.Nil(t, result, "middleware error should cause the event to be skipped (continue)")
@@ -163,7 +163,7 @@ func TestCanonicalStreamConverter_Flush_MiddlewareError_Continue(t *testing.T) {
 	chain.Use(&errorMiddleware{})
 	ctx := NewConversionContext(InterfaceTypeChat)
 
-	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic")
+	converter := NewCanonicalStreamConverterWithMiddleware(decoder, encoder, chain, *ctx, "openai", "anthropic", "")
 	result := converter.Flush()
 
 	assert.Len(t, result, 1)

backend/internal/domain/model.go

@@ -1,8 +1,12 @@
 package domain
 
-import "time"
+import (
+	"time"
 
-// Model 模型领域模型
+	"nex/backend/pkg/modelid"
+)
+
+// Model 模型领域模型（id 为 UUID 自动生成）
 type Model struct {
 	ID         string    `json:"id"`
 	ProviderID string    `json:"provider_id"`
@@ -10,3 +14,8 @@ type Model struct {
 	Enabled    bool      `json:"enabled"`
 	CreatedAt  time.Time `json:"created_at"`
 }
+
+// UnifiedModelID 返回统一模型 ID（格式：provider_id/model_name）
+func (m *Model) UnifiedModelID() string {
+	return modelid.FormatUnifiedModelID(m.ProviderID, m.ModelName)
+}

backend/internal/handler/handler_supplemental_test.go

@@ -113,7 +113,6 @@ func TestModelHandler_CreateModel_Success(t *testing.T) {
 	h := NewModelHandler(&mockModelService{})
 
 	body, _ := json.Marshal(map[string]string{
-		"id":          "m1",
 		"provider_id": "p1",
 		"model_name":  "gpt-4",
 	})
@@ -127,7 +126,7 @@ func TestModelHandler_CreateModel_Success(t *testing.T) {
 
 	var result domain.Model
 	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &result))
-	assert.Equal(t, "m1", result.ID)
+	assert.NotEmpty(t, result.ID)
 }
 
 func TestModelHandler_GetModel(t *testing.T) {

backend/internal/handler/handler_test.go

@@ -13,7 +13,6 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"gorm.io/gorm"
 
 	"nex/backend/internal/domain"
 	"nex/backend/internal/provider"
@@ -31,7 +30,7 @@ type mockRoutingService struct {
 	err    error
 }
 
-func (m *mockRoutingService) Route(modelName string) (*domain.RouteResult, error) {
+func (m *mockRoutingService) RouteByModelName(providerID, modelName string) (*domain.RouteResult, error) {
 	return m.result, m.err
 }
 
@@ -57,6 +56,14 @@ type mockProviderService struct {
 	err       error
 }
 
+func (m *mockProviderService) ListEnabledModels() ([]domain.Model, error) {
+	return nil, nil
+}
+
+func (m *mockProviderService) GetModelByProviderAndName(providerID, modelName string) (*domain.Model, error) {
+	return nil, nil
+}
+
 func (m *mockProviderService) Create(provider *domain.Provider) error { return m.err }
 func (m *mockProviderService) Get(id string, maskKey bool) (*domain.Provider, error) {
 	return m.provider, m.err
@@ -73,13 +80,21 @@ type mockModelService struct {
 	err    error
 }
 
-func (m *mockModelService) Create(model *domain.Model) error { return m.err }
+func (m *mockModelService) Create(model *domain.Model) error {
+	if m.err == nil {
+		model.ID = "mock-uuid-1234"
+	}
+	return m.err
+}
 func (m *mockModelService) Get(id string) (*domain.Model, error) {
 	return m.model, m.err
 }
 func (m *mockModelService) List(providerID string) ([]domain.Model, error) {
 	return m.models, m.err
 }
+func (m *mockModelService) ListEnabled() ([]domain.Model, error) {
+	return []domain.Model{}, nil
+}
 func (m *mockModelService) Update(id string, updates map[string]interface{}) error {
 	return m.err
 }
@@ -163,8 +178,8 @@ func TestModelHandler_CreateModel_MissingFields(t *testing.T) {
 func TestModelHandler_ListModels(t *testing.T) {
 	h := NewModelHandler(&mockModelService{
 		models: []domain.Model{
-			{ID: "m1", ModelName: "gpt-4"},
+			{ID: "m1", ProviderID: "openai", ModelName: "gpt-4"},
-			{ID: "m2", ModelName: "gpt-3.5"},
+			{ID: "m2", ProviderID: "anthropic", ModelName: "claude-3"},
 		},
 	})
 
@@ -174,6 +189,72 @@ func TestModelHandler_ListModels(t *testing.T) {
 
 	h.ListModels(c)
 	assert.Equal(t, 200, w.Code)
+
+	var result []modelResponse
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &result))
+	require.Len(t, result, 2)
+	assert.Equal(t, "openai/gpt-4", result[0].UnifiedModelID)
+	assert.Equal(t, "anthropic/claude-3", result[1].UnifiedModelID)
+}
+
+func TestModelHandler_GetModel_UnifiedID(t *testing.T) {
+	h := NewModelHandler(&mockModelService{
+		model: &domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4"},
+	})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "id", Value: "m1"}}
+	c.Request = httptest.NewRequest("GET", "/api/models/m1", nil)
+
+	h.GetModel(c)
+	assert.Equal(t, 200, w.Code)
+
+	var result modelResponse
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &result))
+	assert.Equal(t, "m1", result.ID)
+	assert.Equal(t, "openai/gpt-4", result.UnifiedModelID)
+}
+
+func TestModelHandler_CreateModel_UnifiedID(t *testing.T) {
+	h := NewModelHandler(&mockModelService{})
+
+	body, _ := json.Marshal(map[string]string{
+		"provider_id": "openai",
+		"model_name":  "gpt-4",
+	})
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Request = httptest.NewRequest("POST", "/api/models", bytes.NewReader(body))
+	c.Request.Header.Set("Content-Type", "application/json")
+
+	h.CreateModel(c)
+	assert.Equal(t, 201, w.Code)
+
+	var result modelResponse
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &result))
+	assert.Equal(t, "mock-uuid-1234", result.ID)
+	assert.Equal(t, "openai/gpt-4", result.UnifiedModelID)
+}
+
+func TestModelHandler_UpdateModel_UnifiedID(t *testing.T) {
+	h := NewModelHandler(&mockModelService{
+		model: &domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4-turbo"},
+	})
+
+	body, _ := json.Marshal(map[string]interface{}{"enabled": false})
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "id", Value: "m1"}}
+	c.Request = httptest.NewRequest("PUT", "/api/models/m1", bytes.NewReader(body))
+	c.Request.Header.Set("Content-Type", "application/json")
+
+	h.UpdateModel(c)
+	assert.Equal(t, 200, w.Code)
+
+	var result modelResponse
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &result))
+	assert.Equal(t, "openai/gpt-4-turbo", result.UnifiedModelID)
 }
 
 // ============ Stats Handler 测试 ============
@@ -256,7 +337,7 @@ func formatMapErrors(errs map[string]string) string {
 
 func TestProviderHandler_CreateProvider_DuplicatedKey(t *testing.T) {
 	h := NewProviderHandler(&mockProviderService{
-		err: gorm.ErrDuplicatedKey,
+		err: appErrors.ErrConflict,
 	})
 
 	body, _ := json.Marshal(map[string]string{

backend/internal/handler/model_handler.go

@@ -1,6 +1,7 @@
 package handler
 
 import (
+	"errors"
 	"net/http"
 
 	"github.com/gin-gonic/gin"
@@ -22,23 +23,35 @@ func NewModelHandler(modelService service.ModelService) *ModelHandler {
 	return &ModelHandler{modelService: modelService}
 }
 
+// modelResponse 模型响应 DTO，扩展 unified_id 字段
+type modelResponse struct {
+	domain.Model
+	UnifiedModelID string `json:"unified_id"`
+}
+
+// newModelResponse 从 domain.Model 构造响应 DTO
+func newModelResponse(m *domain.Model) modelResponse {
+	return modelResponse{
+		Model:          *m,
+		UnifiedModelID: m.UnifiedModelID(),
+	}
+}
+
 // CreateModel 创建模型
 func (h *ModelHandler) CreateModel(c *gin.Context) {
 	var req struct {
-		ID         string `json:"id" binding:"required"`
 		ProviderID string `json:"provider_id" binding:"required"`
 		ModelName  string `json:"model_name" binding:"required"`
 	}
 
 	if err := c.ShouldBindJSON(&req); err != nil {
 		c.JSON(http.StatusBadRequest, gin.H{
-			"error": "缺少必需字段: id, provider_id, model_name",
+			"error": "缺少必需字段: provider_id, model_name",
 		})
 		return
 	}
 
 	model := &domain.Model{
-		ID:         req.ID,
 		ProviderID: req.ProviderID,
 		ModelName:  req.ModelName,
 	}
@@ -51,11 +64,18 @@ func (h *ModelHandler) CreateModel(c *gin.Context) {
 			})
 			return
 		}
+		if err == appErrors.ErrDuplicateModel {
+			c.JSON(http.StatusConflict, gin.H{
+				"error":  "同一供应商下模型名称已存在",
+				"code":   appErrors.ErrDuplicateModel.Code,
+			})
+			return
+		}
 		writeError(c, err)
 		return
 	}
 
-	c.JSON(http.StatusCreated, model)
+	c.JSON(http.StatusCreated, newModelResponse(model))
 }
 
 // ListModels 列出模型
@@ -68,7 +88,11 @@ func (h *ModelHandler) ListModels(c *gin.Context) {
 		return
 	}
 
-	c.JSON(http.StatusOK, models)
+	resp := make([]modelResponse, len(models))
+	for i, m := range models {
+		resp[i] = newModelResponse(&m)
+	}
+	c.JSON(http.StatusOK, resp)
 }
 
 // GetModel 获取模型
@@ -87,7 +111,7 @@ func (h *ModelHandler) GetModel(c *gin.Context) {
 		return
 	}
 
-	c.JSON(http.StatusOK, model)
+	c.JSON(http.StatusOK, newModelResponse(model))
 }
 
 // UpdateModel 更新模型
@@ -104,18 +128,25 @@ func (h *ModelHandler) UpdateModel(c *gin.Context) {
 
 	err := h.modelService.Update(id, req)
 	if err != nil {
-		if err == gorm.ErrRecordNotFound {
+		if errors.Is(err, appErrors.ErrModelNotFound) {
 			c.JSON(http.StatusNotFound, gin.H{
 				"error": "模型未找到",
 			})
 			return
 		}
-		if err == appErrors.ErrProviderNotFound {
+		if errors.Is(err, appErrors.ErrProviderNotFound) {
 			c.JSON(http.StatusBadRequest, gin.H{
 				"error": "供应商不存在",
 			})
 			return
 		}
+		if errors.Is(err, appErrors.ErrDuplicateModel) {
+			c.JSON(http.StatusConflict, gin.H{
+				"error": appErrors.ErrDuplicateModel.Message,
+				"code":  appErrors.ErrDuplicateModel.Code,
+			})
+			return
+		}
 		writeError(c, err)
 		return
 	}
@@ -126,7 +157,7 @@ func (h *ModelHandler) UpdateModel(c *gin.Context) {
 		return
 	}
 
-	c.JSON(http.StatusOK, model)
+	c.JSON(http.StatusOK, newModelResponse(model))
 }
 
 // DeleteModel 删除模型

backend/internal/handler/provider_handler.go

@@ -55,9 +55,10 @@ func (h *ProviderHandler) CreateProvider(c *gin.Context) {
 
 	err := h.providerService.Create(provider)
 	if err != nil {
-		if errors.Is(err, gorm.ErrDuplicatedKey) {
+		if err == appErrors.ErrInvalidProviderID {
-			c.JSON(http.StatusConflict, gin.H{
+			c.JSON(http.StatusBadRequest, gin.H{
-				"error": "供应商 ID 已存在",
+				"error": appErrors.ErrInvalidProviderID.Message,
+				"code":  appErrors.ErrInvalidProviderID.Code,
 			})
 			return
 		}
@@ -119,6 +120,13 @@ func (h *ProviderHandler) UpdateProvider(c *gin.Context) {
 			})
 			return
 		}
+		if errors.Is(err, appErrors.ErrImmutableField) {
+			c.JSON(http.StatusBadRequest, gin.H{
+				"error": appErrors.ErrImmutableField.Message,
+				"code":  appErrors.ErrImmutableField.Code,
+			})
+			return
+		}
 		writeError(c, err)
 		return
 	}

backend/internal/handler/proxy_handler.go

@@ -11,9 +11,11 @@ import (
 	"go.uber.org/zap"
 
 	"nex/backend/internal/conversion"
+	"nex/backend/internal/conversion/canonical"
 	"nex/backend/internal/domain"
 	"nex/backend/internal/provider"
 	"nex/backend/internal/service"
+	"nex/backend/pkg/modelid"
 )
 
 // ProxyHandler 统一代理处理器
@@ -54,6 +56,34 @@ func (h *ProxyHandler) HandleProxy(c *gin.Context) {
 	}
 	nativePath := "/v1/" + path
 
+	// 获取 client adapter
+	registry := h.engine.GetRegistry()
+	clientAdapter, err := registry.Get(clientProtocol)
+	if err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "不支持的协议: " + clientProtocol})
+		return
+	}
+
+	// 检测接口类型
+	ifaceType := clientAdapter.DetectInterfaceType(nativePath)
+
+	// 处理 Models 接口：本地聚合
+	if ifaceType == conversion.InterfaceTypeModels {
+		h.handleModelsList(c, clientAdapter)
+		return
+	}
+
+	// 处理 ModelInfo 接口：本地查询
+	if ifaceType == conversion.InterfaceTypeModelInfo {
+		unifiedID, err := clientAdapter.ExtractUnifiedModelID(nativePath)
+		if err != nil {
+			c.JSON(http.StatusBadRequest, gin.H{"error": "无效的模型 ID 格式"})
+			return
+		}
+		h.handleModelInfo(c, unifiedID, clientAdapter)
+		return
+	}
+
 	// 读取请求体
 	body, err := io.ReadAll(c.Request.Body)
 	if err != nil {
@@ -61,10 +91,17 @@ func (h *ProxyHandler) HandleProxy(c *gin.Context) {
 		return
 	}
 
-	// 解析 model 名称（从 JSON body 中提取，GET 请求无 body）
+	// 解析统一模型 ID（使用 adapter.ExtractModelName）
-	modelName := ""
+	var providerID, modelName string
 	if len(body) > 0 {
-		modelName = extractModelName(body)
+		unifiedID, err := clientAdapter.ExtractModelName(body, ifaceType)
+		if err == nil && unifiedID != "" {
+			pid, mn, err := modelid.ParseUnifiedModelID(unifiedID)
+			if err == nil {
+				providerID = pid
+				modelName = mn
+			}
+		}
 	}
 
 	// 构建输入 HTTPRequestSpec
@@ -76,7 +113,7 @@ func (h *ProxyHandler) HandleProxy(c *gin.Context) {
 	}
 
 	// 路由
-	routeResult, err := h.routingService.Route(modelName)
+	routeResult, err := h.routingService.RouteByModelName(providerID, modelName)
 	if err != nil {
 		// GET 请求或无法提取 model 时，直接转发到上游
 		if len(body) == 0 || modelName == "" {
@@ -94,24 +131,30 @@ func (h *ProxyHandler) HandleProxy(c *gin.Context) {
 	}
 
 	// 构建 TargetProvider
+	// 注意：ModelName 字段用于 Smart Passthrough 场景改写请求体
+	// 同协议：请求体中的统一 ID 会被改写为 ModelName（上游名）
+	// 跨协议：全量转换时 ModelName 会被编码到请求体中
 	targetProvider := conversion.NewTargetProvider(
 		routeResult.Provider.BaseURL,
 		routeResult.Provider.APIKey,
-		routeResult.Model.ModelName,
+		routeResult.Model.ModelName,  // 上游模型名，用于请求改写
 	)
 
 	// 判断是否流式
 	isStream := h.isStreamRequest(body, clientProtocol, nativePath)
 
+	// 计算统一模型 ID（用于响应覆写）
+	unifiedModelID := routeResult.Model.UnifiedModelID()
+
 	if isStream {
-		h.handleStream(c, inSpec, clientProtocol, providerProtocol, targetProvider, routeResult)
+		h.handleStream(c, inSpec, clientProtocol, providerProtocol, targetProvider, routeResult, unifiedModelID, ifaceType)
 	} else {
-		h.handleNonStream(c, inSpec, clientProtocol, providerProtocol, targetProvider, routeResult)
+		h.handleNonStream(c, inSpec, clientProtocol, providerProtocol, targetProvider, routeResult, unifiedModelID, ifaceType)
 	}
 }
 
 // handleNonStream 处理非流式请求
-func (h *ProxyHandler) handleNonStream(c *gin.Context, inSpec conversion.HTTPRequestSpec, clientProtocol, providerProtocol string, targetProvider *conversion.TargetProvider, routeResult *domain.RouteResult) {
+func (h *ProxyHandler) handleNonStream(c *gin.Context, inSpec conversion.HTTPRequestSpec, clientProtocol, providerProtocol string, targetProvider *conversion.TargetProvider, routeResult *domain.RouteResult, unifiedModelID string, ifaceType conversion.InterfaceType) {
 	// 转换请求
 	outSpec, err := h.engine.ConvertHttpRequest(inSpec, clientProtocol, providerProtocol, targetProvider)
 	if err != nil {
@@ -128,9 +171,8 @@ func (h *ProxyHandler) handleNonStream(c *gin.Context, inSpec conversion.HTTPReq
 		return
 	}
 
-	// 转换响应
+	// 转换响应，传入 modelOverride（跨协议场景覆写 model 字段）
-	interfaceType, _ := h.engine.DetectInterfaceType(inSpec.URL, clientProtocol)
+	convertedResp, err := h.engine.ConvertHttpResponse(*resp, clientProtocol, providerProtocol, ifaceType, unifiedModelID)
-	convertedResp, err := h.engine.ConvertHttpResponse(*resp, clientProtocol, providerProtocol, interfaceType)
 	if err != nil {
 		h.logger.Error("转换响应失败", zap.String("error", err.Error()))
 		h.writeConversionError(c, err, clientProtocol)
@@ -153,7 +195,7 @@ func (h *ProxyHandler) handleNonStream(c *gin.Context, inSpec conversion.HTTPReq
 }
 
 // handleStream 处理流式请求
-func (h *ProxyHandler) handleStream(c *gin.Context, inSpec conversion.HTTPRequestSpec, clientProtocol, providerProtocol string, targetProvider *conversion.TargetProvider, routeResult *domain.RouteResult) {
+func (h *ProxyHandler) handleStream(c *gin.Context, inSpec conversion.HTTPRequestSpec, clientProtocol, providerProtocol string, targetProvider *conversion.TargetProvider, routeResult *domain.RouteResult, unifiedModelID string, ifaceType conversion.InterfaceType) {
 	// 转换请求
 	outSpec, err := h.engine.ConvertHttpRequest(inSpec, clientProtocol, providerProtocol, targetProvider)
 	if err != nil {
@@ -161,8 +203,8 @@ func (h *ProxyHandler) handleStream(c *gin.Context, inSpec conversion.HTTPReques
 		return
 	}
 
-	// 创建流式转换器
+	// 创建流式转换器，传入 modelOverride（跨协议场景覆写 model 字段）
-	streamConverter, err := h.engine.CreateStreamConverter(clientProtocol, providerProtocol)
+	streamConverter, err := h.engine.CreateStreamConverter(clientProtocol, providerProtocol, unifiedModelID, ifaceType)
 	if err != nil {
 		h.writeConversionError(c, err, clientProtocol)
 		return
@@ -224,6 +266,79 @@ func (h *ProxyHandler) isStreamRequest(body []byte, clientProtocol, nativePath s
 	return req.Stream
 }
 
+// handleModelsList 处理 GET /v1/models 本地聚合
+func (h *ProxyHandler) handleModelsList(c *gin.Context, adapter conversion.ProtocolAdapter) {
+	// 从数据库查询所有启用的模型
+	models, err := h.providerService.ListEnabledModels()
+	if err != nil {
+		h.logger.Error("查询启用模型失败", zap.String("error", err.Error()))
+		c.JSON(http.StatusInternalServerError, gin.H{"error": "查询模型失败"})
+		return
+	}
+
+	// 构建 CanonicalModelList
+	modelList := &canonical.CanonicalModelList{
+		Models: make([]canonical.CanonicalModel, 0, len(models)),
+	}
+
+	for _, m := range models {
+		modelList.Models = append(modelList.Models, canonical.CanonicalModel{
+			ID:      m.UnifiedModelID(),
+			Name:    m.ModelName,
+			Created: m.CreatedAt.Unix(),
+			OwnedBy: m.ProviderID,
+		})
+	}
+
+	// 使用 adapter 编码返回
+	body, err := adapter.EncodeModelsResponse(modelList)
+	if err != nil {
+		h.logger.Error("编码 Models 响应失败", zap.String("error", err.Error()))
+		c.JSON(http.StatusInternalServerError, gin.H{"error": "编码响应失败"})
+		return
+	}
+
+	c.Data(http.StatusOK, "application/json", body)
+}
+
+// handleModelInfo 处理 GET /v1/models/{unified_id} 本地查询
+func (h *ProxyHandler) handleModelInfo(c *gin.Context, unifiedID string, adapter conversion.ProtocolAdapter) {
+	// 解析统一模型 ID
+	providerID, modelName, err := modelid.ParseUnifiedModelID(unifiedID)
+	if err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{
+			"error": "无效的统一模型 ID 格式",
+			"code":  "INVALID_MODEL_ID",
+		})
+		return
+	}
+
+	// 从数据库查询模型
+	model, err := h.providerService.GetModelByProviderAndName(providerID, modelName)
+	if err != nil {
+		c.JSON(http.StatusNotFound, gin.H{"error": "模型未找到"})
+		return
+	}
+
+	// 构建 CanonicalModelInfo
+	modelInfo := &canonical.CanonicalModelInfo{
+		ID:      model.UnifiedModelID(),
+		Name:    model.ModelName,
+		Created: model.CreatedAt.Unix(),
+		OwnedBy: model.ProviderID,
+	}
+
+	// 使用 adapter 编码返回
+	body, err := adapter.EncodeModelInfoResponse(modelInfo)
+	if err != nil {
+		h.logger.Error("编码 ModelInfo 响应失败", zap.String("error", err.Error()))
+		c.JSON(http.StatusInternalServerError, gin.H{"error": "编码响应失败"})
+		return
+	}
+
+	c.Data(http.StatusOK, "application/json", body)
+}
+
 // writeConversionError 写入转换错误
 func (h *ProxyHandler) writeConversionError(c *gin.Context, err error, clientProtocol string) {
 	if convErr, ok := err.(*conversion.ConversionError); ok {
@@ -292,7 +407,7 @@ func (h *ProxyHandler) forwardPassthrough(c *gin.Context, inSpec conversion.HTTP
 		return
 	}
 
-	convertedResp, err := h.engine.ConvertHttpResponse(*resp, clientProtocol, providerProtocol, ifaceType)
+	convertedResp, err := h.engine.ConvertHttpResponse(*resp, clientProtocol, providerProtocol, ifaceType, "")
 	if err != nil {
 		h.writeConversionError(c, err, clientProtocol)
 		return
@@ -307,17 +422,6 @@ func (h *ProxyHandler) forwardPassthrough(c *gin.Context, inSpec conversion.HTTP
 	c.Data(convertedResp.StatusCode, "application/json", convertedResp.Body)
 }
 
-// extractModelName 从 JSON body 中提取 model
-func extractModelName(body []byte) string {
-	var req struct {
-		Model string `json:"model"`
-	}
-	if err := json.Unmarshal(body, &req); err != nil {
-		return ""
-	}
-	return req.Model
-}
-
 // extractHeaders 从 Gin context 提取请求头
 func extractHeaders(c *gin.Context) map[string]string {
 	headers := make(map[string]string)

backend/internal/handler/proxy_handler_test.go

@@ -60,13 +60,23 @@ type mockProxyRoutingService struct {
 	err    error
 }
 
-func (m *mockProxyRoutingService) Route(modelName string) (*domain.RouteResult, error) {
+func (m *mockProxyRoutingService) RouteByModelName(providerID, modelName string) (*domain.RouteResult, error) {
 	return m.result, m.err
 }
 
 type mockProxyProviderService struct {
-	providers []domain.Provider
+	providers       []domain.Provider
-	err       error
+	err             error
+	enabledModels   []domain.Model
+	modelByProvName *domain.Model
+}
+
+func (m *mockProxyProviderService) ListEnabledModels() ([]domain.Model, error) {
+	return m.enabledModels, nil
+}
+
+func (m *mockProxyProviderService) GetModelByProviderAndName(providerID, modelName string) (*domain.Model, error) {
+	return m.modelByProvName, nil
 }
 
 func (m *mockProxyProviderService) Create(p *domain.Provider) error                              { return nil }
@@ -319,7 +329,8 @@ func TestProxyHandler_ForwardPassthrough_NoProviders(t *testing.T) {
 	c.Request = httptest.NewRequest("GET", "/openai/v1/models", nil)
 
 	h.HandleProxy(c)
-	assert.Equal(t, 404, w.Code)
+	// Models 接口现在本地聚合，返回空列表 200
+	assert.Equal(t, 200, w.Code)
 }
 
 func TestExtractHeaders(t *testing.T) {
@@ -716,58 +727,6 @@ func TestProxyHandler_HandleProxy_RouteEmptyBody_NoModel(t *testing.T) {
 	assert.Equal(t, 200, w.Code)
 }
 
-// ============ extractModelName 测试 ============
-
-func TestExtractModelName(t *testing.T) {
-	tests := []struct {
-		name     string
-		body     []byte
-		expected string
-	}{
-		{
-			name:     "valid model",
-			body:     []byte(`{"model": "gpt-4", "messages": []}`),
-			expected: "gpt-4",
-		},
-		{
-			name:     "empty body",
-			body:     []byte(`{}`),
-			expected: "",
-		},
-		{
-			name:     "invalid json",
-			body:     []byte(`{invalid}`),
-			expected: "",
-		},
-		{
-			name:     "nested structure",
-			body:     []byte(`{"model": "claude-3", "messages": [{"role": "user", "content": "hello"}]}`),
-			expected: "claude-3",
-		},
-		{
-			name:     "model with special chars",
-			body:     []byte(`{"model": "gpt-4-0125-preview", "stream": true}`),
-			expected: "gpt-4-0125-preview",
-		},
-		{
-			name:     "empty body bytes",
-			body:     []byte{},
-			expected: "",
-		},
-		{
-			name:     "model is null",
-			body:     []byte(`{"model": null}`),
-			expected: "",
-		},
-	}
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			result := extractModelName(tt.body)
-			assert.Equal(t, tt.expected, result)
-		})
-	}
-}
-
 // ============ isStreamRequest 测试 ============
 
 func TestIsStreamRequest(t *testing.T) {
@@ -831,3 +790,270 @@ func TestIsStreamRequest(t *testing.T) {
 		})
 	}
 }
+
+// ============ Models / ModelInfo 本地聚合测试 ============
+
+func TestProxyHandler_HandleProxy_Models_LocalAggregation(t *testing.T) {
+	engine := setupProxyEngine(t)
+	providerSvc := &mockProxyProviderService{
+		enabledModels: []domain.Model{
+			{ID: "m1", ProviderID: "openai", ModelName: "gpt-4", Enabled: true},
+			{ID: "m2", ProviderID: "anthropic", ModelName: "claude-3", Enabled: true},
+		},
+	}
+	h := newTestProxyHandler(engine, &mockProxyProviderClient{}, &mockProxyRoutingService{}, providerSvc)
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/models"}}
+	c.Request = httptest.NewRequest("GET", "/openai/v1/models", nil)
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	data, ok := resp["data"].([]interface{})
+	require.True(t, ok)
+	assert.Len(t, data, 2)
+
+	// 验证统一模型 ID 格式
+	first := data[0].(map[string]interface{})
+	assert.Equal(t, "openai/gpt-4", first["id"])
+}
+
+func TestProxyHandler_HandleProxy_ModelInfo_LocalQuery(t *testing.T) {
+	engine := setupProxyEngine(t)
+	providerSvc := &mockProxyProviderService{
+		modelByProvName: &domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4", Enabled: true},
+	}
+	h := newTestProxyHandler(engine, &mockProxyProviderClient{}, &mockProxyRoutingService{}, providerSvc)
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/models/openai/gpt-4"}}
+	c.Request = httptest.NewRequest("GET", "/openai/v1/models/openai/gpt-4", nil)
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	assert.Equal(t, "openai/gpt-4", resp["id"])
+}
+
+func TestProxyHandler_HandleProxy_Models_EmptySuffix_ForwardPassthrough(t *testing.T) {
+	engine := setupProxyEngine(t)
+	providerSvc := &mockProxyProviderService{
+		providers: []domain.Provider{
+			{ID: "p1", Name: "Test", APIKey: "sk-test", BaseURL: "https://api.test.com", Protocol: "openai"},
+		},
+	}
+	client := &mockProxyProviderClient{
+		sendFn: func(ctx context.Context, spec conversion.HTTPRequestSpec) (*conversion.HTTPResponseSpec, error) {
+			return &conversion.HTTPResponseSpec{
+				StatusCode: 200,
+				Body:       []byte(`{"object":"list","data":[]}`),
+			}, nil
+		},
+	}
+	h := newTestProxyHandler(engine, client, &mockProxyRoutingService{err: appErrors.ErrModelNotFound}, providerSvc)
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/models/"}}
+	c.Request = httptest.NewRequest("GET", "/openai/v1/models/", nil)
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+}
+
+// ============ Smart Passthrough 统一模型 ID 路由测试 ============
+
+func TestProxyHandler_HandleProxy_SmartPassthrough_UnifiedID(t *testing.T) {
+	engine := setupProxyEngine(t)
+	routingSvc := &mockProxyRoutingService{
+		result: &domain.RouteResult{
+			Provider: &domain.Provider{ID: "openai_p", Name: "Test", APIKey: "sk-test", BaseURL: "https://api.test.com", Protocol: "openai", Enabled: true},
+			Model:    &domain.Model{ID: "m1", ProviderID: "openai_p", ModelName: "gpt-4", Enabled: true},
+		},
+	}
+	client := &mockProxyProviderClient{
+		sendFn: func(ctx context.Context, spec conversion.HTTPRequestSpec) (*conversion.HTTPResponseSpec, error) {
+			// 验证请求体中的 model 已被改写为上游模型名
+			var req map[string]interface{}
+			json.Unmarshal(spec.Body, &req)
+			assert.Equal(t, "gpt-4", req["model"])
+
+			return &conversion.HTTPResponseSpec{
+				StatusCode: 200,
+				Headers:    map[string]string{"Content-Type": "application/json"},
+				Body:       []byte(`{"id":"resp-1","object":"chat.completion","model":"gpt-4","choices":[{"index":0,"message":{"role":"assistant","content":"Hello"},"finish_reason":"stop"}],"usage":{"prompt_tokens":5,"completion_tokens":3,"total_tokens":8}}`),
+			}, nil
+		},
+	}
+	h := newTestProxyHandler(engine, client, routingSvc, &mockProxyProviderService{})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/chat/completions"}}
+	// 客户端发送统一模型 ID
+	c.Request = httptest.NewRequest("POST", "/openai/v1/chat/completions", bytes.NewReader([]byte(`{"model":"openai_p/gpt-4","messages":[{"role":"user","content":"hi"}]}`)))
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+
+	// 验证响应中的 model 已被改写为统一模型 ID
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	assert.Equal(t, "openai_p/gpt-4", resp["model"])
+}
+
+// ============ 跨协议统一模型 ID 路由测试 ============
+
+func TestProxyHandler_HandleProxy_CrossProtocol_NonStream_UnifiedID(t *testing.T) {
+	engine := setupProxyEngine(t)
+	routingSvc := &mockProxyRoutingService{
+		result: &domain.RouteResult{
+			Provider: &domain.Provider{ID: "anthropic_p", Name: "Anthropic", APIKey: "sk-test", BaseURL: "https://api.anthropic.com", Protocol: "anthropic", Enabled: true},
+			Model:    &domain.Model{ID: "m1", ProviderID: "anthropic_p", ModelName: "claude-3", Enabled: true},
+		},
+	}
+	client := &mockProxyProviderClient{
+		sendFn: func(ctx context.Context, spec conversion.HTTPRequestSpec) (*conversion.HTTPResponseSpec, error) {
+			return &conversion.HTTPResponseSpec{
+				StatusCode: 200,
+				Headers:    map[string]string{"Content-Type": "application/json"},
+				Body:       []byte(`{"id":"msg-1","type":"message","role":"assistant","model":"claude-3","content":[{"type":"text","text":"Hello"}],"stop_reason":"end_turn","usage":{"input_tokens":5,"output_tokens":3}}`),
+			}, nil
+		},
+	}
+	h := newTestProxyHandler(engine, client, routingSvc, &mockProxyProviderService{})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/chat/completions"}}
+	// OpenAI 客户端使用统一模型 ID 路由到 Anthropic 供应商
+	c.Request = httptest.NewRequest("POST", "/openai/v1/chat/completions", bytes.NewReader([]byte(`{"model":"anthropic_p/claude-3","messages":[{"role":"user","content":"hi"}]}`)))
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+
+	// 验证跨协议转换后响应中的 model 被覆写为统一模型 ID
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	assert.Equal(t, "anthropic_p/claude-3", resp["model"])
+}
+
+func TestProxyHandler_HandleProxy_CrossProtocol_Stream_UnifiedID(t *testing.T) {
+	engine := setupProxyEngine(t)
+	routingSvc := &mockProxyRoutingService{
+		result: &domain.RouteResult{
+			Provider: &domain.Provider{ID: "anthropic_p", Name: "Anthropic", APIKey: "sk-test", BaseURL: "https://api.anthropic.com", Protocol: "anthropic", Enabled: true},
+			Model:    &domain.Model{ID: "m1", ProviderID: "anthropic_p", ModelName: "claude-3", Enabled: true},
+		},
+	}
+	client := &mockProxyProviderClient{
+		sendStreamFn: func(ctx context.Context, spec conversion.HTTPRequestSpec) (<-chan provider.StreamEvent, error) {
+			ch := make(chan provider.StreamEvent, 10)
+			go func() {
+				defer close(ch)
+				ch <- provider.StreamEvent{Data: []byte(`event: message_start
+data: {"type":"message_start","message":{"id":"msg-1","type":"message","role":"assistant","model":"claude-3","content":[]}}
+
+`)}
+				ch <- provider.StreamEvent{Data: []byte(`event: content_block_delta
+data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"Hi"}}
+
+`)}
+				ch <- provider.StreamEvent{Data: []byte(`event: message_stop
+data: {"type":"message_stop"}
+
+`)}
+				ch <- provider.StreamEvent{Done: true}
+			}()
+			return ch, nil
+		},
+	}
+	h := newTestProxyHandler(engine, client, routingSvc, &mockProxyProviderService{})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/chat/completions"}}
+	c.Request = httptest.NewRequest("POST", "/openai/v1/chat/completions", bytes.NewReader([]byte(`{"model":"anthropic_p/claude-3","messages":[{"role":"user","content":"hi"}],"stream":true}`)))
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+	assert.Equal(t, "text/event-stream", w.Header().Get("Content-Type"))
+
+	body := w.Body.String()
+	// 验证跨协议流式中 model 被覆写为统一模型 ID
+	assert.Contains(t, body, "anthropic_p/claude-3", "跨协议流式响应中 model 应被覆写为统一模型 ID")
+}
+
+func TestProxyHandler_HandleProxy_SmartPassthrough_Fidelity(t *testing.T) {
+	engine := setupProxyEngine(t)
+	routingSvc := &mockProxyRoutingService{
+		result: &domain.RouteResult{
+			Provider: &domain.Provider{ID: "openai_p", Name: "Test", APIKey: "sk-test", BaseURL: "https://api.test.com", Protocol: "openai", Enabled: true},
+			Model:    &domain.Model{ID: "m1", ProviderID: "openai_p", ModelName: "gpt-4", Enabled: true},
+		},
+	}
+	var capturedRequestBody []byte
+	client := &mockProxyProviderClient{
+		sendFn: func(ctx context.Context, spec conversion.HTTPRequestSpec) (*conversion.HTTPResponseSpec, error) {
+			capturedRequestBody = spec.Body
+			return &conversion.HTTPResponseSpec{
+				StatusCode: 200,
+				Headers:    map[string]string{"Content-Type": "application/json"},
+				Body:       []byte(`{"id":"resp-1","object":"chat.completion","model":"gpt-4","choices":[{"index":0,"message":{"role":"assistant","content":"Hello"},"finish_reason":"stop"}],"usage":{"prompt_tokens":5,"completion_tokens":3,"total_tokens":8},"unknown_field":"preserved"}`),
+			}, nil
+		},
+	}
+	h := newTestProxyHandler(engine, client, routingSvc, &mockProxyProviderService{})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/chat/completions"}}
+	// 包含未知参数，验证 Smart Passthrough 保真性
+	c.Request = httptest.NewRequest("POST", "/openai/v1/chat/completions", bytes.NewReader([]byte(`{"model":"openai_p/gpt-4","messages":[{"role":"user","content":"hi"}],"custom_param":"should_be_preserved"}`)))
+
+	h.HandleProxy(c)
+	assert.Equal(t, 200, w.Code)
+
+	// 验证请求中 model 被改写为上游模型名，但未知参数保留
+	var reqBody map[string]interface{}
+	require.NoError(t, json.Unmarshal(capturedRequestBody, &reqBody))
+	assert.Equal(t, "gpt-4", reqBody["model"], "请求中 model 应被改写为上游模型名")
+	assert.Equal(t, "should_be_preserved", reqBody["custom_param"], "Smart Passthrough 应保留未知参数")
+
+	// 验证响应中 model 被改写为统一模型 ID，但未知参数保留
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	assert.Equal(t, "openai_p/gpt-4", resp["model"], "响应中 model 应被改写为统一模型 ID")
+	assert.Equal(t, "preserved", resp["unknown_field"], "Smart Passthrough 应保留未知响应字段")
+}
+
+func TestProxyHandler_HandleProxy_UnifiedID_ModelNotFound(t *testing.T) {
+	engine := setupProxyEngine(t)
+	routingSvc := &mockProxyRoutingService{err: appErrors.ErrModelNotFound}
+	providerSvc := &mockProxyProviderService{
+		providers: []domain.Provider{
+			{ID: "p1", Name: "Test", APIKey: "sk-test", BaseURL: "https://api.test.com", Protocol: "openai"},
+		},
+	}
+	h := newTestProxyHandler(engine, &mockProxyProviderClient{}, routingSvc, providerSvc)
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+	c.Params = gin.Params{{Key: "protocol", Value: "openai"}, {Key: "path", Value: "/chat/completions"}}
+	// 使用统一模型 ID 格式但模型不存在
+	c.Request = httptest.NewRequest("POST", "/openai/v1/chat/completions", bytes.NewReader([]byte(`{"model":"unknown/model","messages":[{"role":"user","content":"hi"}]}`)))
+
+	h.HandleProxy(c)
+	assert.Equal(t, 404, w.Code)
+
+	var resp map[string]interface{}
+	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
+	assert.Contains(t, resp, "error")
+}

backend/internal/repository/model_repo.go

@@ -7,7 +7,8 @@ type ModelRepository interface {
 	Create(model *domain.Model) error
 	GetByID(id string) (*domain.Model, error)
 	List(providerID string) ([]domain.Model, error)
-	GetByModelName(modelName string) (*domain.Model, error)
+	FindByProviderAndModelName(providerID, modelName string) (*domain.Model, error)
+	ListEnabled() ([]domain.Model, error)
 	Update(id string, updates map[string]interface{}) error
 	Delete(id string) error
 }

backend/internal/repository/model_repo_impl.go

@@ -52,9 +52,9 @@ func (r *modelRepository) List(providerID string) ([]domain.Model, error) {
 	return result, nil
 }
 
-func (r *modelRepository) GetByModelName(modelName string) (*domain.Model, error) {
+func (r *modelRepository) FindByProviderAndModelName(providerID, modelName string) (*domain.Model, error) {
 	var m config.Model
-	err := r.db.Where("model_name = ?", modelName).First(&m).Error
+	err := r.db.Where("provider_id = ? AND model_name = ?", providerID, modelName).First(&m).Error
 	if err != nil {
 		return nil, err
 	}
@@ -62,6 +62,21 @@ func (r *modelRepository) GetByModelName(modelName string) (*domain.Model, error
 	return &d, nil
 }
 
+func (r *modelRepository) ListEnabled() ([]domain.Model, error) {
+	var models []config.Model
+	err := r.db.Joins("JOIN providers ON providers.id = models.provider_id").
+		Where("models.enabled = ? AND providers.enabled = ?", true, true).
+		Find(&models).Error
+	if err != nil {
+		return nil, err
+	}
+	result := make([]domain.Model, len(models))
+	for i := range models {
+		result[i] = toDomainModel(&models[i])
+	}
+	return result, nil
+}
+
 func (r *modelRepository) Update(id string, updates map[string]interface{}) error {
 	result := r.db.Model(&config.Model{}).Where("id = ?", id).Updates(updates)
 	if result.Error != nil {

backend/internal/repository/provider_repo.go

@@ -9,4 +9,7 @@ type ProviderRepository interface {
 	List() ([]domain.Provider, error)
 	Update(id string, updates map[string]interface{}) error
 	Delete(id string) error
+	// 统一模型 ID 相关方法
+	ListEnabledModels() ([]domain.Model, error)
+	FindByProviderAndModelName(providerID, modelName string) (*domain.Model, error)
 }

backend/internal/repository/provider_repo_impl.go

@@ -71,6 +71,25 @@ func (r *providerRepository) Delete(id string) error {
 	return nil
 }
 
+// ListEnabledModels 返回所有启用的模型（关联启用的供应商）
+func (r *providerRepository) ListEnabledModels() ([]domain.Model, error) {
+	var models []domain.Model
+	err := r.db.Joins("JOIN providers ON providers.id = models.provider_id").
+		Where("models.enabled = ? AND providers.enabled = ?", true, true).
+		Find(&models).Error
+	return models, err
+}
+
+// FindByProviderAndModelName 按 provider_id 和 model_name 查询模型
+func (r *providerRepository) FindByProviderAndModelName(providerID, modelName string) (*domain.Model, error) {
+	var model domain.Model
+	err := r.db.Where("provider_id = ? AND model_name = ?", providerID, modelName).First(&model).Error
+	if err != nil {
+		return nil, err
+	}
+	return &model, nil
+}
+
 func toDomainProvider(p *config.Provider) domain.Provider {
 	return domain.Provider{
 		ID:        p.ID,

backend/internal/repository/repository_test.go

@@ -147,15 +147,36 @@ func TestModelRepository_GetByID(t *testing.T) {
 	assert.Equal(t, "gpt-4", result.ModelName)
 }
 
-func TestModelRepository_GetByModelName(t *testing.T) {
+func TestModelRepository_FindByProviderAndModelName(t *testing.T) {
 	db := setupTestDB(t)
 	repo := NewModelRepository(db)
 
 	repo.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4", Enabled: true})
 
-	result, err := repo.GetByModelName("gpt-4")
+	result, err := repo.FindByProviderAndModelName("p1", "gpt-4")
 	require.NoError(t, err)
 	assert.Equal(t, "m1", result.ID)
+	assert.Equal(t, "p1", result.ProviderID)
+	assert.Equal(t, "gpt-4", result.ModelName)
+}
+
+func TestModelRepository_FindByProviderAndModelName_NotFound(t *testing.T) {
+	db := setupTestDB(t)
+	repo := NewModelRepository(db)
+
+	repo.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4", Enabled: true})
+
+	// Wrong provider_id
+	_, err := repo.FindByProviderAndModelName("p2", "gpt-4")
+	assert.Error(t, err)
+
+	// Wrong model_name
+	_, err = repo.FindByProviderAndModelName("p1", "gpt-3.5")
+	assert.Error(t, err)
+
+	// Both wrong
+	_, err = repo.FindByProviderAndModelName("p2", "claude-3")
+	assert.Error(t, err)
 }
 
 func TestModelRepository_List(t *testing.T) {
@@ -175,6 +196,54 @@ func TestModelRepository_List(t *testing.T) {
 	assert.Len(t, p1Models, 2)
 }
 
+func TestModelRepository_ListEnabled(t *testing.T) {
+	db := setupTestDB(t)
+	providerRepo := NewProviderRepository(db)
+	modelRepo := NewModelRepository(db)
+
+	// Create two providers (both start enabled due to gorm:"default:true")
+	err := providerRepo.Create(&domain.Provider{
+		ID: "enabled-provider", Name: "Enabled Provider",
+		APIKey: "key1", BaseURL: "https://enabled.com", Enabled: true,
+	})
+	require.NoError(t, err)
+	err = providerRepo.Create(&domain.Provider{
+		ID: "disabled-provider", Name: "Disabled Provider",
+		APIKey: "key2", BaseURL: "https://disabled.com", Enabled: true,
+	})
+	require.NoError(t, err)
+
+	// Disable the second provider via Update (GORM default:true skips zero values on Create)
+	err = providerRepo.Update("disabled-provider", map[string]interface{}{"enabled": false})
+	require.NoError(t, err)
+
+	// Create models (all start enabled due to gorm:"default:true")
+	err = modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "enabled-provider", ModelName: "gpt-4", Enabled: true})
+	require.NoError(t, err)
+	err = modelRepo.Create(&domain.Model{ID: "m2", ProviderID: "enabled-provider", ModelName: "gpt-3.5", Enabled: true})
+	require.NoError(t, err)
+	err = modelRepo.Create(&domain.Model{ID: "m3", ProviderID: "disabled-provider", ModelName: "claude-3", Enabled: true})
+	require.NoError(t, err)
+	err = modelRepo.Create(&domain.Model{ID: "m4", ProviderID: "disabled-provider", ModelName: "claude-3.5", Enabled: true})
+	require.NoError(t, err)
+
+	// Disable m2 via Update
+	err = modelRepo.Update("m2", map[string]interface{}{"enabled": false})
+	require.NoError(t, err)
+
+	// ListEnabled should only return models where both model and provider are enabled:
+	// - m1: enabled provider + enabled model -> returned
+	// - m2: enabled provider + disabled model -> filtered out
+	// - m3: disabled provider + enabled model -> filtered out
+	// - m4: disabled provider + enabled model -> filtered out
+	enabled, err := modelRepo.ListEnabled()
+	require.NoError(t, err)
+	require.Len(t, enabled, 1)
+	assert.Equal(t, "m1", enabled[0].ID)
+	assert.Equal(t, "enabled-provider", enabled[0].ProviderID)
+	assert.Equal(t, "gpt-4", enabled[0].ModelName)
+}
+
 func TestModelRepository_Update(t *testing.T) {
 	db := setupTestDB(t)
 	repo := NewModelRepository(db)

backend/internal/service/model_service.go

@@ -7,6 +7,7 @@ type ModelService interface {
 	Create(model *domain.Model) error
 	Get(id string) (*domain.Model, error)
 	List(providerID string) ([]domain.Model, error)
+	ListEnabled() ([]domain.Model, error)
 	Update(id string, updates map[string]interface{}) error
 	Delete(id string) error
 }

backend/internal/service/model_service_impl.go

@@ -1,6 +1,7 @@
 package service
 
 import (
+	"github.com/google/uuid"
 	appErrors "nex/backend/pkg/errors"
 
 	"nex/backend/internal/domain"
@@ -17,11 +18,18 @@ func NewModelService(modelRepo repository.ModelRepository, providerRepo reposito
 }
 
 func (s *modelService) Create(model *domain.Model) error {
-	// Verify provider exists
+	// 校验供应商存在
-	_, err := s.providerRepo.GetByID(model.ProviderID)
+	if _, err := s.providerRepo.GetByID(model.ProviderID); err != nil {
-	if err != nil {
 		return appErrors.ErrProviderNotFound
 	}
+
+	// 联合唯一校验：同一供应商下 model_name 不重复
+	if err := s.checkDuplicateModelName(model.ProviderID, model.ModelName, ""); err != nil {
+		return err
+	}
+
+	// 自动生成 UUID 作为 id
+	model.ID = uuid.New().String()
 	model.Enabled = true
 	return s.modelRepo.Create(model)
 }
@@ -34,17 +42,57 @@ func (s *modelService) List(providerID string) ([]domain.Model, error) {
 	return s.modelRepo.List(providerID)
 }
 
+func (s *modelService) ListEnabled() ([]domain.Model, error) {
+	return s.modelRepo.ListEnabled()
+}
+
 func (s *modelService) Update(id string, updates map[string]interface{}) error {
-	// If updating provider_id, verify new provider exists
+	// 获取当前模型
+	current, err := s.modelRepo.GetByID(id)
+	if err != nil {
+		return appErrors.ErrModelNotFound
+	}
+
+	// 如果更新 provider_id，校验新供应商存在
 	if providerID, ok := updates["provider_id"].(string); ok {
-		_, err := s.providerRepo.GetByID(providerID)
+		if _, err := s.providerRepo.GetByID(providerID); err != nil {
-		if err != nil {
 			return appErrors.ErrProviderNotFound
 		}
 	}
+
+	// 确定更新后的 provider_id 和 model_name
+	newProviderID := current.ProviderID
+	if v, ok := updates["provider_id"].(string); ok {
+		newProviderID = v
+	}
+	newModelName := current.ModelName
+	if v, ok := updates["model_name"].(string); ok {
+		newModelName = v
+	}
+
+	// 如果 provider_id 或 model_name 发生变化，校验联合唯一
+	if newProviderID != current.ProviderID || newModelName != current.ModelName {
+		if err := s.checkDuplicateModelName(newProviderID, newModelName, id); err != nil {
+			return err
+		}
+	}
+
 	return s.modelRepo.Update(id, updates)
 }
 
 func (s *modelService) Delete(id string) error {
 	return s.modelRepo.Delete(id)
 }
+
+// checkDuplicateModelName 校验同一供应商下 model_name 是否重复
+// excludeID 用于更新时排除自身
+func (s *modelService) checkDuplicateModelName(providerID, modelName, excludeID string) error {
+	existing, err := s.modelRepo.FindByProviderAndModelName(providerID, modelName)
+	if err != nil {
+		return nil // 未找到，不重复
+	}
+	if excludeID != "" && existing.ID == excludeID {
+		return nil // 排除自身
+	}
+	return appErrors.ErrDuplicateModel
+}

backend/internal/service/provider_service.go

@@ -9,4 +9,7 @@ type ProviderService interface {
 	List() ([]domain.Provider, error)
 	Update(id string, updates map[string]interface{}) error
 	Delete(id string) error
+	// 统一模型 ID 相关方法
+	ListEnabledModels() ([]domain.Model, error)
+	GetModelByProviderAndName(providerID, modelName string) (*domain.Model, error)
 }

backend/internal/service/provider_service_impl.go

@@ -1,21 +1,35 @@
 package service
 
 import (
+	"strings"
+
+	"nex/backend/pkg/modelid"
+
 	"nex/backend/internal/domain"
 	"nex/backend/internal/repository"
+	appErrors "nex/backend/pkg/errors"
 )
 
 type providerService struct {
 	providerRepo repository.ProviderRepository
+	modelRepo    repository.ModelRepository
 }
 
-func NewProviderService(providerRepo repository.ProviderRepository) ProviderService {
+func NewProviderService(providerRepo repository.ProviderRepository, modelRepo repository.ModelRepository) ProviderService {
-	return &providerService{providerRepo: providerRepo}
+	return &providerService{providerRepo: providerRepo, modelRepo: modelRepo}
 }
 
 func (s *providerService) Create(provider *domain.Provider) error {
+	// 校验 provider_id 字符集
+	if err := modelid.ValidateProviderID(provider.ID); err != nil {
+		return appErrors.ErrInvalidProviderID
+	}
 	provider.Enabled = true
-	return s.providerRepo.Create(provider)
+	err := s.providerRepo.Create(provider)
+	if err != nil && isUniqueConstraintError(err) {
+		return appErrors.ErrConflict
+	}
+	return err
 }
 
 func (s *providerService) Get(id string, maskKey bool) (*domain.Provider, error) {
@@ -41,9 +55,31 @@ func (s *providerService) List() ([]domain.Provider, error) {
 }
 
 func (s *providerService) Update(id string, updates map[string]interface{}) error {
+	if _, ok := updates["id"]; ok {
+		return appErrors.ErrImmutableField
+	}
 	return s.providerRepo.Update(id, updates)
 }
 
 func (s *providerService) Delete(id string) error {
 	return s.providerRepo.Delete(id)
 }
+
+// ListEnabledModels 返回所有启用的模型（用于 Models 接口本地聚合）
+func (s *providerService) ListEnabledModels() ([]domain.Model, error) {
+	return s.modelRepo.ListEnabled()
+}
+
+// GetModelByProviderAndName 按 provider_id 和 model_name 查询模型（用于 ModelInfo 接口本地查询）
+func (s *providerService) GetModelByProviderAndName(providerID, modelName string) (*domain.Model, error) {
+	return s.modelRepo.FindByProviderAndModelName(providerID, modelName)
+}
+
+// isUniqueConstraintError 判断是否为数据库唯一约束冲突错误
+func isUniqueConstraintError(err error) bool {
+	if err == nil {
+		return false
+	}
+	msg := strings.ToLower(err.Error())
+	return strings.Contains(msg, "unique constraint") || strings.Contains(msg, "duplicate")
+}

backend/internal/service/routing_service.go

@@ -4,5 +4,5 @@ import "nex/backend/internal/domain"
 
 // RoutingService 路由服务接口
 type RoutingService interface {
-	Route(modelName string) (*domain.RouteResult, error)
+	RouteByModelName(providerID, modelName string) (*domain.RouteResult, error)
 }

backend/internal/service/routing_service_impl.go

@@ -16,8 +16,8 @@ func NewRoutingService(modelRepo repository.ModelRepository, providerRepo reposi
 	return &routingService{modelRepo: modelRepo, providerRepo: providerRepo}
 }
 
-func (s *routingService) Route(modelName string) (*domain.RouteResult, error) {
+func (s *routingService) RouteByModelName(providerID, modelName string) (*domain.RouteResult, error) {
-	model, err := s.modelRepo.GetByModelName(modelName)
+	model, err := s.modelRepo.FindByProviderAndModelName(providerID, modelName)
 	if err != nil {
 		return nil, appErrors.ErrModelNotFound
 	}

backend/internal/service/service_supplemental_test.go

@@ -13,7 +13,8 @@ import (
 func TestProviderService_Update(t *testing.T) {
 	db := setupServiceTestDB(t)
 	repo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewProviderService(repo, modelRepo)
 
 	svc.Create(&domain.Provider{ID: "p1", Name: "Original", APIKey: "key", BaseURL: "https://test.com"})
 
@@ -28,7 +29,8 @@ func TestProviderService_Update(t *testing.T) {
 func TestProviderService_Update_NotFound(t *testing.T) {
 	db := setupServiceTestDB(t)
 	repo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewProviderService(repo, modelRepo)
 
 	err := svc.Update("nonexistent", map[string]interface{}{"name": "test"})
 	assert.Error(t, err)
@@ -41,11 +43,12 @@ func TestModelService_Get(t *testing.T) {
 	svc := NewModelService(modelRepo, providerRepo)
 
 	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
-	svc.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"})
+	model := &domain.Model{ProviderID: "p1", ModelName: "gpt-4"}
+	require.NoError(t, svc.Create(model))
 
-	model, err := svc.Get("m1")
+	result, err := svc.Get(model.ID)
 	require.NoError(t, err)
-	assert.Equal(t, "gpt-4", model.ModelName)
+	assert.Equal(t, "gpt-4", result.ModelName)
 }
 
 func TestModelService_Update(t *testing.T) {
@@ -55,14 +58,15 @@ func TestModelService_Update(t *testing.T) {
 	svc := NewModelService(modelRepo, providerRepo)
 
 	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
-	svc.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"})
+	model := &domain.Model{ProviderID: "p1", ModelName: "gpt-4"}
+	require.NoError(t, svc.Create(model))
 
-	err := svc.Update("m1", map[string]interface{}{"model_name": "gpt-4o"})
+	err := svc.Update(model.ID, map[string]interface{}{"model_name": "gpt-4o"})
 	require.NoError(t, err)
 
-	model, err := svc.Get("m1")
+	result, err := svc.Get(model.ID)
 	require.NoError(t, err)
-	assert.Equal(t, "gpt-4o", model.ModelName)
+	assert.Equal(t, "gpt-4o", result.ModelName)
 }
 
 func TestModelService_Update_ProviderID_Invalid(t *testing.T) {
@@ -72,9 +76,10 @@ func TestModelService_Update_ProviderID_Invalid(t *testing.T) {
 	svc := NewModelService(modelRepo, providerRepo)
 
 	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
-	svc.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"})
+	model := &domain.Model{ProviderID: "p1", ModelName: "gpt-4"}
+	require.NoError(t, svc.Create(model))
 
-	err := svc.Update("m1", map[string]interface{}{"provider_id": "nonexistent"})
+	err := svc.Update(model.ID, map[string]interface{}{"provider_id": "nonexistent"})
 	assert.Error(t, err)
 }
 
@@ -85,12 +90,13 @@ func TestModelService_Delete(t *testing.T) {
 	svc := NewModelService(modelRepo, providerRepo)
 
 	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
-	svc.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"})
+	model := &domain.Model{ProviderID: "p1", ModelName: "gpt-4"}
+	require.NoError(t, svc.Create(model))
 
-	err := svc.Delete("m1")
+	err := svc.Delete(model.ID)
 	require.NoError(t, err)
 
-	_, err = svc.Get("m1")
+	_, err = svc.Get(model.ID)
 	assert.Error(t, err)
 }
 

backend/internal/service/service_test.go

@@ -1,8 +1,10 @@
 package service
 
 import (
+	"errors"
 	"testing"
 
+	"github.com/google/uuid"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 	"gorm.io/driver/sqlite"
@@ -11,6 +13,7 @@ import (
 	"nex/backend/internal/config"
 	"nex/backend/internal/domain"
 	"nex/backend/internal/repository"
+	appErrors "nex/backend/pkg/errors"
 )
 
 func setupServiceTestDB(t *testing.T) *gorm.DB {
@@ -29,80 +32,106 @@ func setupServiceTestDB(t *testing.T) *gorm.DB {
 	return db
 }
 
-// ============ ProviderService 测试 ============
+// ============ RoutingService - RouteByModelName 测试 ============
 
-func TestProviderService_Create(t *testing.T) {
+func TestRoutingService_RouteByModelName_Success(t *testing.T) {
 	db := setupServiceTestDB(t)
-	repo := repository.NewProviderRepository(db)
+	providerRepo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewRoutingService(modelRepo, providerRepo)
 
-	provider := &domain.Provider{
+	// 创建供应商和模型
-		ID: "test-p", Name: "Test", APIKey: "sk-test", BaseURL: "https://api.test.com",
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com", Enabled: true})
-	}
+	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4", Enabled: true})
-	err := svc.Create(provider)
+
+	result, err := svc.RouteByModelName("openai", "gpt-4")
 	require.NoError(t, err)
-	assert.True(t, provider.Enabled)
+	assert.Equal(t, "openai", result.Provider.ID)
+	assert.Equal(t, "gpt-4", result.Model.ModelName)
 }
 
-func TestProviderService_Get_MaskKey(t *testing.T) {
+func TestRoutingService_RouteByModelName_NotFound(t *testing.T) {
 	db := setupServiceTestDB(t)
-	repo := repository.NewProviderRepository(db)
+	providerRepo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewRoutingService(modelRepo, providerRepo)
 
-	svc.Create(&domain.Provider{
+	_, err := svc.RouteByModelName("openai", "nonexistent-model")
-		ID: "p1", Name: "Test", APIKey: "sk-long-api-key-12345", BaseURL: "https://test.com",
+	assert.True(t, errors.Is(err, appErrors.ErrModelNotFound))
-	})
-
-	result, err := svc.Get("p1", true)
-	require.NoError(t, err)
-	assert.Equal(t, "***2345", result.APIKey)
-
-	result, err = svc.Get("p1", false)
-	require.NoError(t, err)
-	assert.Equal(t, "sk-long-api-key-12345", result.APIKey)
 }
 
-func TestProviderService_List(t *testing.T) {
+func TestRoutingService_RouteByModelName_DisabledModel(t *testing.T) {
 	db := setupServiceTestDB(t)
-	repo := repository.NewProviderRepository(db)
+	providerRepo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewRoutingService(modelRepo, providerRepo)
 
-	svc.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key123", BaseURL: "https://a.com"})
+	// 创建启用的供应商和禁用的模型
-	svc.Create(&domain.Provider{ID: "p2", Name: "P2", APIKey: "key456", BaseURL: "https://b.com"})
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com", Enabled: true})
+	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4", Enabled: true})
+	modelRepo.Update("m1", map[string]interface{}{"enabled": false})
 
-	providers, err := svc.List()
+	_, err := svc.RouteByModelName("openai", "gpt-4")
-	require.NoError(t, err)
+	assert.True(t, errors.Is(err, appErrors.ErrModelDisabled))
-	assert.Len(t, providers, 2)
-	assert.Contains(t, providers[0].APIKey, "***")
 }
 
-func TestProviderService_Delete(t *testing.T) {
+func TestRoutingService_RouteByModelName_DisabledProvider(t *testing.T) {
 	db := setupServiceTestDB(t)
-	repo := repository.NewProviderRepository(db)
+	providerRepo := repository.NewProviderRepository(db)
-	svc := NewProviderService(repo)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewRoutingService(modelRepo, providerRepo)
 
-	svc.Create(&domain.Provider{ID: "p1", Name: "Test", APIKey: "key", BaseURL: "https://test.com"})
+	// 创建启用的供应商和模型，然后禁用供应商
-	err := svc.Delete("p1")
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com", Enabled: true})
-	require.NoError(t, err)
+	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "openai", ModelName: "gpt-4", Enabled: true})
+	providerRepo.Update("openai", map[string]interface{}{"enabled": false})
 
-	_, err = svc.Get("p1", false)
+	_, err := svc.RouteByModelName("openai", "gpt-4")
-	assert.Error(t, err)
+	assert.True(t, errors.Is(err, appErrors.ErrProviderDisabled))
 }
 
-// ============ ModelService 测试 ============
+// ============ ModelService - Create with UUID 测试 ============
 
-func TestModelService_Create(t *testing.T) {
+func TestModelService_Create_GeneratesUUID(t *testing.T) {
 	db := setupServiceTestDB(t)
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
 	svc := NewModelService(modelRepo, providerRepo)
 
-	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"})
 
-	model := &domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"}
+	model := &domain.Model{ProviderID: "openai", ModelName: "gpt-4"}
 	err := svc.Create(model)
 	require.NoError(t, err)
-	assert.True(t, model.Enabled)
+
+	// 验证返回的 model 拥有有效的 UUID
+	assert.NotEmpty(t, model.ID)
+	_, err = uuid.Parse(model.ID)
+	assert.NoError(t, err, "model.ID should be a valid UUID")
+
+	// 通过 Get 验证持久化
+	stored, err := svc.Get(model.ID)
+	require.NoError(t, err)
+	assert.Equal(t, model.ID, stored.ID)
+	assert.Equal(t, "gpt-4", stored.ModelName)
+}
+
+func TestModelService_Create_DuplicateModelName(t *testing.T) {
+	db := setupServiceTestDB(t)
+	providerRepo := repository.NewProviderRepository(db)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewModelService(modelRepo, providerRepo)
+
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"})
+
+	model1 := &domain.Model{ProviderID: "openai", ModelName: "gpt-4"}
+	err := svc.Create(model1)
+	require.NoError(t, err)
+
+	// 使用相同的 (providerID, modelName) 创建第二个模型应失败
+	model2 := &domain.Model{ProviderID: "openai", ModelName: "gpt-4"}
+	err = svc.Create(model2)
+	assert.True(t, errors.Is(err, appErrors.ErrDuplicateModel))
 }
 
 func TestModelService_Create_ProviderNotFound(t *testing.T) {
@@ -111,160 +140,135 @@ func TestModelService_Create_ProviderNotFound(t *testing.T) {
 	modelRepo := repository.NewModelRepository(db)
 	svc := NewModelService(modelRepo, providerRepo)
 
-	model := &domain.Model{ID: "m1", ProviderID: "nonexistent", ModelName: "gpt-4"}
+	model := &domain.Model{ProviderID: "nonexistent", ModelName: "gpt-4"}
 	err := svc.Create(model)
-	assert.Error(t, err)
+	assert.True(t, errors.Is(err, appErrors.ErrProviderNotFound))
 }
 
-func TestModelService_List(t *testing.T) {
+// ============ ProviderService - Create with validation 测试 ============
+
+func TestProviderService_Create_InvalidID(t *testing.T) {
+	db := setupServiceTestDB(t)
+	repo := repository.NewProviderRepository(db)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewProviderService(repo, modelRepo)
+
+	provider := &domain.Provider{ID: "open-ai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"}
+	err := svc.Create(provider)
+	assert.True(t, errors.Is(err, appErrors.ErrInvalidProviderID))
+}
+
+func TestProviderService_Create_ValidID(t *testing.T) {
+	db := setupServiceTestDB(t)
+	repo := repository.NewProviderRepository(db)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewProviderService(repo, modelRepo)
+
+	provider := &domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"}
+	err := svc.Create(provider)
+	require.NoError(t, err)
+	assert.Equal(t, "openai", provider.ID)
+	assert.True(t, provider.Enabled)
+}
+
+// ============ ModelService - Update with duplicate check 测试 ============
+
+func TestModelService_Update_DuplicateModelName(t *testing.T) {
 	db := setupServiceTestDB(t)
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
 	svc := NewModelService(modelRepo, providerRepo)
 
-	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com"})
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"})
-	svc.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4"})
+	providerRepo.Create(&domain.Provider{ID: "anthropic", Name: "Anthropic", APIKey: "key", BaseURL: "https://api.anthropic.com"})
-	svc.Create(&domain.Model{ID: "m2", ProviderID: "p1", ModelName: "gpt-3.5"})
 
-	models, err := svc.List("p1")
+	model1 := &domain.Model{ProviderID: "openai", ModelName: "gpt-4"}
+	err := svc.Create(model1)
 	require.NoError(t, err)
-	assert.Len(t, models, 2)
+
+	model2 := &domain.Model{ProviderID: "anthropic", ModelName: "claude-3"}
+	err = svc.Create(model2)
+	require.NoError(t, err)
+
+	// 将 model2 的 model_name 改为 "gpt-4" 且 provider_id 改为 "openai"，与 model1 冲突
+	err = svc.Update(model2.ID, map[string]interface{}{
+		"provider_id": "openai",
+		"model_name":  "gpt-4",
+	})
+	assert.True(t, errors.Is(err, appErrors.ErrDuplicateModel))
 }
 
-// ============ RoutingService 测试 ============
+func TestModelService_Update_ModelNotFound(t *testing.T) {
-
-func TestRoutingService_Route(t *testing.T) {
 	db := setupServiceTestDB(t)
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
-	svc := NewRoutingService(modelRepo, providerRepo)
+	svc := NewModelService(modelRepo, providerRepo)
 
-	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com", Enabled: true})
+	err := svc.Update("nonexistent-id", map[string]interface{}{
-	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4", Enabled: true})
+		"model_name": "gpt-4",
-
+	})
-	result, err := svc.Route("gpt-4")
+	assert.True(t, errors.Is(err, appErrors.ErrModelNotFound))
-	require.NoError(t, err)
-	assert.Equal(t, "p1", result.Provider.ID)
-	assert.Equal(t, "gpt-4", result.Model.ModelName)
 }
 
-func TestRoutingService_Route_ModelNotFound(t *testing.T) {
+func TestModelService_Update_Success(t *testing.T) {
 	db := setupServiceTestDB(t)
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
-	svc := NewRoutingService(modelRepo, providerRepo)
+	svc := NewModelService(modelRepo, providerRepo)
 
-	_, err := svc.Route("nonexistent-model")
+	providerRepo.Create(&domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"})
-	assert.Error(t, err)
-}
 
-func TestRoutingService_Route_ModelDisabled(t *testing.T) {
+	model := &domain.Model{ProviderID: "openai", ModelName: "gpt-4"}
-	db := setupServiceTestDB(t)
+	err := svc.Create(model)
-	providerRepo := repository.NewProviderRepository(db)
-	modelRepo := repository.NewModelRepository(db)
-	svc := NewRoutingService(modelRepo, providerRepo)
-
-	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com", Enabled: true})
-	// 先创建启用的模型，然后通过 Update 禁用
-	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4", Enabled: true})
-	modelRepo.Update("m1", map[string]interface{}{"enabled": false})
-
-	_, err := svc.Route("gpt-4")
-	assert.Error(t, err)
-}
-
-func TestRoutingService_Route_ProviderDisabled(t *testing.T) {
-	db := setupServiceTestDB(t)
-	providerRepo := repository.NewProviderRepository(db)
-	modelRepo := repository.NewModelRepository(db)
-	svc := NewRoutingService(modelRepo, providerRepo)
-
-	// 先创建启用的 provider，然后禁用
-	providerRepo.Create(&domain.Provider{ID: "p1", Name: "P1", APIKey: "key", BaseURL: "https://test.com", Enabled: true})
-	providerRepo.Update("p1", map[string]interface{}{"enabled": false})
-	modelRepo.Create(&domain.Model{ID: "m1", ProviderID: "p1", ModelName: "gpt-4", Enabled: true})
-
-	_, err := svc.Route("gpt-4")
-	assert.Error(t, err)
-}
-
-// ============ StatsService 测试 ============
-
-func TestStatsService_RecordAndGet(t *testing.T) {
-	db := setupServiceTestDB(t)
-	statsRepo := repository.NewStatsRepository(db)
-	svc := NewStatsService(statsRepo)
-
-	err := svc.Record("p1", "gpt-4")
 	require.NoError(t, err)
 
-	stats, err := svc.Get("p1", "", nil, nil)
+	// 更新 model_name 为不冲突的值
+	err = svc.Update(model.ID, map[string]interface{}{
+		"model_name": "gpt-4-turbo",
+	})
 	require.NoError(t, err)
-	assert.Len(t, stats, 1)
+
+	updated, err := svc.Get(model.ID)
+	require.NoError(t, err)
+	assert.Equal(t, "gpt-4-turbo", updated.ModelName)
 }
 
-func TestStatsService_Aggregate_ByProvider(t *testing.T) {
+// ============ ProviderService - Update immutable ID 测试 ============
-	statsRepo := repository.NewStatsRepository(nil)
-	svc := NewStatsService(statsRepo)
 
-	stats := []domain.UsageStats{
+func TestProviderService_Update_ImmutableID(t *testing.T) {
-		{ProviderID: "p1", ModelName: "gpt-4", RequestCount: 10},
+	db := setupServiceTestDB(t)
-		{ProviderID: "p1", ModelName: "gpt-3.5", RequestCount: 5},
+	repo := repository.NewProviderRepository(db)
-		{ProviderID: "p2", ModelName: "claude-3", RequestCount: 8},
+	modelRepo := repository.NewModelRepository(db)
-	}
+	svc := NewProviderService(repo, modelRepo)
 
-	result := svc.Aggregate(stats, "provider")
+	provider := &domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"}
-	assert.Len(t, result, 2)
+	err := svc.Create(provider)
+	require.NoError(t, err)
 
-	p1Count := 0
+	// 尝试更新 id 字段
-	p2Count := 0
+	err = svc.Update("openai", map[string]interface{}{
-	for _, r := range result {
+		"id": "new-id",
-		if r["provider_id"] == "p1" {
+	})
-			p1Count = r["request_count"].(int)
+	assert.True(t, errors.Is(err, appErrors.ErrImmutableField))
-		}
-		if r["provider_id"] == "p2" {
-			p2Count = r["request_count"].(int)
-		}
-	}
-	assert.Equal(t, 15, p1Count)
-	assert.Equal(t, 8, p2Count)
 }
 
-func TestStatsService_Aggregate_ByDate(t *testing.T) {
+func TestProviderService_Update_Success(t *testing.T) {
-	statsRepo := repository.NewStatsRepository(nil)
+	db := setupServiceTestDB(t)
-	svc := NewStatsService(statsRepo)
+	repo := repository.NewProviderRepository(db)
+	modelRepo := repository.NewModelRepository(db)
+	svc := NewProviderService(repo, modelRepo)
 
-	stats := []domain.UsageStats{
+	provider := &domain.Provider{ID: "openai", Name: "OpenAI", APIKey: "key", BaseURL: "https://api.openai.com"}
-		{ProviderID: "p1", RequestCount: 10},
+	err := svc.Create(provider)
-		{ProviderID: "p2", RequestCount: 5},
+	require.NoError(t, err)
-	}
 
-	result := svc.Aggregate(stats, "date")
+	// 更新 name
-	assert.Len(t, result, 1)
+	err = svc.Update("openai", map[string]interface{}{
-	assert.Equal(t, 15, result[0]["request_count"])
+		"name": "OpenAI Updated",
-}
+	})
-
+	require.NoError(t, err)
-func TestStatsService_Aggregate_ByModel(t *testing.T) {
+
-	statsRepo := repository.NewStatsRepository(nil)
+	updated, err := svc.Get("openai", false)
-	svc := NewStatsService(statsRepo)
+	require.NoError(t, err)
-
+	assert.Equal(t, "OpenAI Updated", updated.Name)
-	stats := []domain.UsageStats{
-		{ProviderID: "openai", ModelName: "gpt-4", RequestCount: 10},
-		{ProviderID: "openai", ModelName: "gpt-3.5", RequestCount: 5},
-		{ProviderID: "anthropic", ModelName: "claude-3", RequestCount: 8},
-		{ProviderID: "openai", ModelName: "gpt-4", RequestCount: 3},
-	}
-
-	result := svc.Aggregate(stats, "model")
-	assert.Len(t, result, 3)
-
-	// 验证每个 provider/model 组合的计数
-	counts := make(map[string]int)
-	for _, r := range result {
-		key := r["provider_id"].(string) + "/" + r["model_name"].(string)
-		counts[key] = r["request_count"].(int)
-	}
-	assert.Equal(t, 13, counts["openai/gpt-4"])
-	assert.Equal(t, 5, counts["openai/gpt-3.5"])
-	assert.Equal(t, 8, counts["anthropic/claude-3"])
 }

backend/migrations/20260401000002_add_indexes.sql

@@ -1,9 +0,0 @@
--- +goose Up
-CREATE INDEX IF NOT EXISTS idx_models_provider_id ON models(provider_id);
-CREATE INDEX IF NOT EXISTS idx_models_model_name ON models(model_name);
-CREATE INDEX IF NOT EXISTS idx_usage_stats_provider_model_date ON usage_stats(provider_id, model_name, date);
-
--- +goose Down
-DROP INDEX IF EXISTS idx_usage_stats_provider_model_date;
-DROP INDEX IF EXISTS idx_models_model_name;
-DROP INDEX IF EXISTS idx_models_provider_id;

backend/migrations/20260419000001_add_provider_protocol.sql

@@ -1,6 +0,0 @@
--- +goose Up
-ALTER TABLE providers ADD COLUMN protocol TEXT DEFAULT 'openai';
-
--- +goose Down
--- SQLite 不支持 DROP COLUMN（3.35.0 之前），但 goose 的 Down 通常不需要
-CREATE TABLE providers_backup AS SELECT id, name, api_key, base_url, enabled, created_at, updated_at FROM providers;

backend/migrations/20260421000001_initial_schema.sql

@@ -1,9 +1,13 @@
 -- +goose Up
+-- 统一初始迁移：providers、models、usage_stats 完整表结构
+-- models 表使用 UUID 主键 + UNIQUE(provider_id, model_name) 联合唯一约束
+
 CREATE TABLE IF NOT EXISTS providers (
     id TEXT PRIMARY KEY,
     name TEXT NOT NULL,
     api_key TEXT NOT NULL,
     base_url TEXT NOT NULL,
+    protocol TEXT DEFAULT 'openai',
     enabled INTEGER DEFAULT 1,
     created_at DATETIME,
     updated_at DATETIME
@@ -15,7 +19,8 @@ CREATE TABLE IF NOT EXISTS models (
     model_name TEXT NOT NULL,
     enabled INTEGER DEFAULT 1,
     created_at DATETIME,
-    FOREIGN KEY (provider_id) REFERENCES providers(id) ON DELETE CASCADE
+    FOREIGN KEY (provider_id) REFERENCES providers(id) ON DELETE CASCADE,
+    UNIQUE(provider_id, model_name)
 );
 
 CREATE TABLE IF NOT EXISTS usage_stats (
@@ -27,7 +32,14 @@ CREATE TABLE IF NOT EXISTS usage_stats (
     UNIQUE(provider_id, model_name, date)
 );
 
+CREATE INDEX IF NOT EXISTS idx_models_provider_id ON models(provider_id);
+CREATE INDEX IF NOT EXISTS idx_models_model_name ON models(model_name);
+CREATE INDEX IF NOT EXISTS idx_usage_stats_provider_model_date ON usage_stats(provider_id, model_name, date);
+
 -- +goose Down
+DROP INDEX IF EXISTS idx_usage_stats_provider_model_date;
+DROP INDEX IF EXISTS idx_models_model_name;
+DROP INDEX IF EXISTS idx_models_provider_id;
 DROP TABLE IF EXISTS usage_stats;
 DROP TABLE IF EXISTS models;
 DROP TABLE IF EXISTS providers;

backend/pkg/errors/errors.go

@@ -49,17 +49,20 @@ func NewAppError(code, message string, httpStatus int) *AppError {
 
 // Predefined errors
 var (
-	ErrModelNotFound    = NewAppError("model_not_found", "模型未找到", http.StatusNotFound)
+	ErrModelNotFound     = NewAppError("model_not_found", "模型未找到", http.StatusNotFound)
-	ErrModelDisabled    = NewAppError("model_disabled", "模型已禁用", http.StatusNotFound)
+	ErrModelDisabled     = NewAppError("model_disabled", "模型已禁用", http.StatusNotFound)
-	ErrProviderNotFound = NewAppError("provider_not_found", "供应商未找到", http.StatusNotFound)
+	ErrProviderNotFound  = NewAppError("provider_not_found", "供应商未找到", http.StatusNotFound)
-	ErrProviderDisabled = NewAppError("provider_disabled", "供应商已禁用", http.StatusNotFound)
+	ErrProviderDisabled  = NewAppError("provider_disabled", "供应商已禁用", http.StatusNotFound)
-	ErrInvalidRequest   = NewAppError("invalid_request", "无效的请求", http.StatusBadRequest)
+	ErrInvalidRequest    = NewAppError("invalid_request", "无效的请求", http.StatusBadRequest)
-	ErrInternal         = NewAppError("internal_error", "内部错误", http.StatusInternalServerError)
+	ErrInternal          = NewAppError("internal_error", "内部错误", http.StatusInternalServerError)
-	ErrDatabaseNotInit  = NewAppError("database_not_initialized", "数据库未初始化", http.StatusInternalServerError)
+	ErrDatabaseNotInit   = NewAppError("database_not_initialized", "数据库未初始化", http.StatusInternalServerError)
-	ErrConflict         = NewAppError("conflict", "资源已存在", http.StatusConflict)
+	ErrConflict          = NewAppError("conflict", "资源已存在", http.StatusConflict)
-	ErrRequestCreate    = NewAppError("request_create_error", "创建请求失败", http.StatusInternalServerError)
+	ErrRequestCreate     = NewAppError("request_create_error", "创建请求失败", http.StatusInternalServerError)
-	ErrRequestSend      = NewAppError("request_send_error", "发送请求失败", http.StatusBadGateway)
+	ErrRequestSend       = NewAppError("request_send_error", "发送请求失败", http.StatusBadGateway)
-	ErrResponseRead     = NewAppError("response_read_error", "读取响应失败", http.StatusBadGateway)
+	ErrResponseRead      = NewAppError("response_read_error", "读取响应失败", http.StatusBadGateway)
+	ErrInvalidProviderID = NewAppError("invalid_provider_id", "供应商 ID 仅允许字母、数字、下划线，长度 1-64", http.StatusBadRequest)
+	ErrDuplicateModel    = NewAppError("duplicate_model", "同一供应商下模型名称已存在", http.StatusConflict)
+	ErrImmutableField    = NewAppError("immutable_field", "供应商 ID 不允许修改", http.StatusBadRequest)
 )
 
 // AsAppError 尝试将 error 转换为 *AppError

backend/pkg/logger/logger.go

@@ -9,6 +9,14 @@ import (
 	"go.uber.org/zap/zapcore"
 )
 
+// stdoutWriter 包装 os.Stdout，忽略 Sync() 错误。
+// 在非 TTY 环境（如 go test）中，os.Stdout 被重定向为 pipe，
+// 底层 fsync 会返回 "bad file descriptor"。zap 社区标准做法。
+type stdoutWriter struct{}
+
+func (stdoutWriter) Write(p []byte) (int, error) { return os.Stdout.Write(p) }
+func (stdoutWriter) Sync() error                 { return nil }
+
 // Config 日志配置
 type Config struct {
 	Level      string // 日志级别: debug, info, warn, error
@@ -46,7 +54,7 @@ func New(cfg Config) (*zap.Logger, error) {
 
 	stdoutCore := zapcore.NewCore(
 		stdoutEncoder,
-		zapcore.AddSync(os.Stdout),
+		zapcore.AddSync(stdoutWriter{}),
 		level,
 	)
 

backend/pkg/modelid/model_id.go

@@ -0,0 +1,63 @@
+package modelid
+
+import (
+	"errors"
+	"regexp"
+	"strings"
+)
+
+var providerIDRegex = regexp.MustCompile(`^[a-zA-Z0-9_]+$`)
+
+// ParseUnifiedModelID 将 "provider_id/model_name" 格式的字符串解析为 providerID 和 modelName
+// 在第一个 "/" 处分割，model_name 可以包含 "/"
+func ParseUnifiedModelID(id string) (providerID, modelName string, err error) {
+	if id == "" {
+		return "", "", errors.New("统一模型 ID 不能为空")
+	}
+
+	parts := strings.SplitN(id, "/", 2)
+	if len(parts) != 2 {
+		return "", "", errors.New("统一模型 ID 格式错误，缺少分隔符 \"/\"")
+	}
+
+	providerID = parts[0]
+	modelName = parts[1]
+
+	if providerID == "" {
+		return "", "", errors.New("provider_id 不能为空")
+	}
+	if modelName == "" {
+		return "", "", errors.New("model_name 不能为空")
+	}
+
+	if !providerIDRegex.MatchString(providerID) {
+		return "", "", errors.New("provider_id 仅允许字母、数字、下划线")
+	}
+
+	return providerID, modelName, nil
+}
+
+// FormatUnifiedModelID 将 providerID 和 modelName 组合格式化为统一模型 ID
+func FormatUnifiedModelID(providerID, modelName string) string {
+	return providerID + "/" + modelName
+}
+
+// ValidateProviderID 校验 providerID 仅包含字母、数字、下划线，长度 1-64
+func ValidateProviderID(id string) error {
+	if id == "" {
+		return errors.New("provider_id 不能为空")
+	}
+	if len(id) > 64 {
+		return errors.New("provider_id 长度不能超过 64 个字符")
+	}
+	if !providerIDRegex.MatchString(id) {
+		return errors.New("provider_id 仅允许字母、数字、下划线")
+	}
+	return nil
+}
+
+// IsValidUnifiedModelID 判断字符串是否为合法的统一模型 ID 格式
+func IsValidUnifiedModelID(id string) bool {
+	_, _, err := ParseUnifiedModelID(id)
+	return err == nil
+}

backend/pkg/modelid/model_id_test.go

@@ -0,0 +1,96 @@
+package modelid
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestParseUnifiedModelID_StandardFormat(t *testing.T) {
+	providerID, modelName, err := ParseUnifiedModelID("openai/gpt-4")
+	assert.NoError(t, err)
+	assert.Equal(t, "openai", providerID)
+	assert.Equal(t, "gpt-4", modelName)
+}
+
+func TestParseUnifiedModelID_ModelNameWithSlashes(t *testing.T) {
+	providerID, modelName, err := ParseUnifiedModelID("azure/accounts/org-123/models/gpt-4")
+	assert.NoError(t, err)
+	assert.Equal(t, "azure", providerID)
+	assert.Equal(t, "accounts/org-123/models/gpt-4", modelName)
+}
+
+func TestParseUnifiedModelID_MissingSeparator(t *testing.T) {
+	_, _, err := ParseUnifiedModelID("gpt-4")
+	assert.Error(t, err)
+}
+
+func TestParseUnifiedModelID_EmptyString(t *testing.T) {
+	_, _, err := ParseUnifiedModelID("")
+	assert.Error(t, err)
+}
+
+func TestParseUnifiedModelID_OnlySeparator(t *testing.T) {
+	tests := []string{"/model", "provider/", "/"}
+	for _, tc := range tests {
+		_, _, err := ParseUnifiedModelID(tc)
+		assert.Error(t, err, "输入 %q 应返回错误", tc)
+	}
+}
+
+func TestParseUnifiedModelID_InvalidProviderID(t *testing.T) {
+	tests := []string{
+		"open-ai/gpt-4",
+		"open.ai/gpt-4",
+		"供应商/gpt-4",
+		"open ai/gpt-4",
+	}
+	for _, tc := range tests {
+		_, _, err := ParseUnifiedModelID(tc)
+		assert.Error(t, err, "providerID 含非法字符 %q 应返回错误", tc)
+	}
+}
+
+func TestFormatUnifiedModelID(t *testing.T) {
+	assert.Equal(t, "openai/gpt-4", FormatUnifiedModelID("openai", "gpt-4"))
+	assert.Equal(t, "anthropic/claude-3-opus", FormatUnifiedModelID("anthropic", "claude-3-opus"))
+}
+
+func TestValidateProviderID_Valid(t *testing.T) {
+	validIDs := []string{"openai", "deep_seek", "provider01", "OpenAI"}
+	for _, id := range validIDs {
+		assert.NoError(t, ValidateProviderID(id), "%q 应校验通过", id)
+	}
+}
+
+func TestValidateProviderID_InvalidChars(t *testing.T) {
+	invalidIDs := []string{"open-ai", "open.ai", "open ai", "供应商", "open/ai"}
+	for _, id := range invalidIDs {
+		assert.Error(t, ValidateProviderID(id), "%q 应校验失败", id)
+	}
+}
+
+func TestValidateProviderID_Empty(t *testing.T) {
+	assert.Error(t, ValidateProviderID(""))
+}
+
+func TestValidateProviderID_TooLong(t *testing.T) {
+	longID := strings.Repeat("a", 65)
+	assert.Error(t, ValidateProviderID(longID))
+
+	exactly64 := strings.Repeat("a", 64)
+	assert.NoError(t, ValidateProviderID(exactly64))
+}
+
+func TestIsValidUnifiedModelID(t *testing.T) {
+	assert.True(t, IsValidUnifiedModelID("openai/gpt-4"))
+	assert.True(t, IsValidUnifiedModelID("anthropic/claude-3-opus-20240229"))
+	assert.True(t, IsValidUnifiedModelID("azure/accounts/org/models/gpt-4"))
+
+	assert.False(t, IsValidUnifiedModelID(""))
+	assert.False(t, IsValidUnifiedModelID("gpt-4"))
+	assert.False(t, IsValidUnifiedModelID("open-ai/gpt-4"))
+	assert.False(t, IsValidUnifiedModelID("/model"))
+	assert.False(t, IsValidUnifiedModelID("provider/"))
+}

backend/tests/helpers.go

@@ -3,6 +3,7 @@ package tests
 import (
 	"fmt"
 	"testing"
+	"time"
 
 	"nex/backend/internal/config"
 
@@ -11,26 +12,36 @@ import (
 	"gorm.io/gorm"
 )
 
-// SetupTestDB initializes a temporary SQLite database with auto-migration.
+// SetupTestDB initializes an in-memory SQLite database with auto-migration.
+// Uses :memory: mode with MaxOpenConns(1) to ensure all operations share the
+// same connection, avoiding "database is closed" errors from connection pool.
+// Enables foreign key constraints for SQLite.
 func SetupTestDB(t *testing.T) *gorm.DB {
 	t.Helper()
 
-	dir := t.TempDir()
+	db, err := gorm.Open(sqlite.Open(":memory:?_foreign_keys=on"), &gorm.Config{})
-	dsn := dir + "/test.db"
-
-	db, err := gorm.Open(sqlite.Open(dsn), &gorm.Config{})
 	assert.NoError(t, err, "failed to open test database")
 
+	// 限制为单连接，确保 :memory: 数据库不被连接池丢弃
+	sqlDB, err := db.DB()
+	assert.NoError(t, err, "failed to get underlying sql.DB")
+	sqlDB.SetMaxOpenConns(1)
+	sqlDB.SetConnMaxLifetime(0)
+
 	err = db.AutoMigrate(&config.Provider{}, &config.Model{}, &config.UsageStats{})
 	assert.NoError(t, err, "failed to auto-migrate test database")
 
 	return db
 }
 
-// CleanupTestDB closes the database and removes the temp database file.
+// CleanupTestDB closes the database after a brief delay to allow async
+// goroutines (e.g. stats recording) to finish.
 func CleanupTestDB(t *testing.T, db *gorm.DB) {
 	t.Helper()
 
+	// 等待异步 goroutine（如 statsService.Record）完成
+	time.Sleep(50 * time.Millisecond)
+
 	sqlDB, err := db.DB()
 	assert.NoError(t, err, "failed to get underlying sql.DB")
 
@@ -57,7 +68,8 @@ func CreateTestProvider(t *testing.T, db *gorm.DB, id string) config.Provider {
 }
 
 // CreateTestModel creates a test model and returns it.
-func CreateTestModel(t *testing.T, db *gorm.DB, id string, providerID string, modelName string) config.Model {
+// Does NOT assert on error - returns the model and error for caller to verify.
+func CreateTestModel(t *testing.T, db *gorm.DB, id string, providerID string, modelName string) (config.Model, error) {
 	t.Helper()
 
 	model := config.Model{
@@ -68,7 +80,5 @@ func CreateTestModel(t *testing.T, db *gorm.DB, id string, providerID string, mo
 	}
 
 	err := db.Create(&model).Error
-	assert.NoError(t, err, "failed to create test model")
+	return model, err
-
-	return model
 }

backend/tests/integration/conversion_test.go

@@ -14,10 +14,8 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"gorm.io/driver/sqlite"
 	"gorm.io/gorm"
 
-	"nex/backend/internal/config"
 	"nex/backend/internal/conversion"
 	"nex/backend/internal/conversion/anthropic"
 	openaiConv "nex/backend/internal/conversion/openai"
@@ -43,11 +41,7 @@ func setupConversionTest(t *testing.T) (*gin.Engine, *gorm.DB, *httptest.Server)
 		w.Write([]byte(`{"error":"not mocked"}`))
 	}))
 
-	dir := t.TempDir()
+	db := setupTestDB(t)
-	db, err := gorm.Open(sqlite.Open(dir+"/test.db"), &gorm.Config{})
-	require.NoError(t, err)
-	err = db.AutoMigrate(&config.Provider{}, &config.Model{}, &config.UsageStats{})
-	require.NoError(t, err)
 	t.Cleanup(func() {
 		sqlDB, _ := db.DB()
 		if sqlDB != nil {
@@ -60,7 +54,7 @@ func setupConversionTest(t *testing.T) (*gin.Engine, *gorm.DB, *httptest.Server)
 	modelRepo := repository.NewModelRepository(db)
 	statsRepo := repository.NewStatsRepository(db)
 
-	providerService := service.NewProviderService(providerRepo)
+	providerService := service.NewProviderService(providerRepo, modelRepo)
 	modelService := service.NewModelService(modelRepo, providerRepo)
 	routingService := service.NewRoutingService(modelRepo, providerRepo)
 	statsService := service.NewStatsService(statsRepo)
@@ -125,7 +119,7 @@ func createProviderAndModel(t *testing.T, r *gin.Engine, providerID, protocol, m
 	require.Equal(t, 201, w.Code)
 
 	modelBody, _ := json.Marshal(map[string]string{
-		"id":          modelName,
+		
 		"provider_id": providerID,
 		"model_name":  modelName,
 	})
@@ -156,7 +150,7 @@ func TestConversion_OpenAIToAnthropic_NonStream(t *testing.T) {
 			"id":    "msg_test",
 			"type":  "message",
 			"role":  "assistant",
-			"model": "claude-3-opus",
+			"model": "anthropic_p/claude-3-opus",
 			"content": []map[string]any{
 				{"type": "text", "text": "Hello from Anthropic!"},
 			},
@@ -170,11 +164,11 @@ func TestConversion_OpenAIToAnthropic_NonStream(t *testing.T) {
 		json.NewEncoder(w).Encode(resp)
 	})
 
-	createProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-3-opus", upstream.URL)
+	createProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-3-opus", upstream.URL)
 
 	// 使用 OpenAI 格式发送请求
 	openaiReq := map[string]any{
-		"model": "claude-3-opus",
+		"model": "anthropic_p/claude-3-opus",
 		"messages": []map[string]any{
 			{"role": "user", "content": "Hello"},
 		},
@@ -233,10 +227,10 @@ func TestConversion_AnthropicToOpenAI_NonStream(t *testing.T) {
 		json.NewEncoder(w).Encode(resp)
 	})
 
-	createProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	createProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	anthropicReq := map[string]any{
-		"model":      "gpt-4",
+		"model":      "openai_p/gpt-4",
 		"max_tokens": 1024,
 		"messages": []map[string]any{
 			{"role": "user", "content": "Hello"},
@@ -273,16 +267,18 @@ func TestConversion_OpenAIToOpenAI_Passthrough(t *testing.T) {
 		body, _ := io.ReadAll(r.Body)
 		var req map[string]any
 		json.Unmarshal(body, &req)
+		// Smart Passthrough: 请求体中的统一 ID 应被改写为上游模型名
 		assert.Equal(t, "gpt-4", req["model"])
 
 		w.Header().Set("Content-Type", "application/json")
+		// 上游返回上游模型名
 		w.Write([]byte(`{"id":"chatcmpl-pass","object":"chat.completion","model":"gpt-4","choices":[{"index":0,"message":{"role":"assistant","content":"passthrough"},"finish_reason":"stop"}],"usage":{"prompt_tokens":5,"completion_tokens":1,"total_tokens":6}}`))
 	})
 
-	createProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	createProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	reqBody := map[string]any{
-		"model":    "gpt-4",
+		"model":    "openai_p/gpt-4",  // 客户端发送统一 ID
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 	}
 	body, _ := json.Marshal(reqBody)
@@ -293,7 +289,8 @@ func TestConversion_OpenAIToOpenAI_Passthrough(t *testing.T) {
 	r.ServeHTTP(w, req)
 
 	assert.Equal(t, 200, w.Code)
-	assert.Contains(t, w.Body.String(), "passthrough")
+	// Smart Passthrough: 响应体中的上游模型名应被改写为统一 ID
+	assert.Contains(t, w.Body.String(), `"model":"openai_p/gpt-4"`)
 }
 
 func TestConversion_AnthropicToAnthropic_Passthrough(t *testing.T) {
@@ -302,14 +299,21 @@ func TestConversion_AnthropicToAnthropic_Passthrough(t *testing.T) {
 	upstream.Config.Handler = http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
 		assert.Equal(t, "/v1/messages", r.URL.Path)
 
+		body, _ := io.ReadAll(r.Body)
+		var req map[string]any
+		json.Unmarshal(body, &req)
+		// Smart Passthrough: 请求体中的统一 ID 应被改写为上游模型名
+		assert.Equal(t, "claude-3-opus", req["model"])
+
+		// 上游返回上游模型名
 		w.Header().Set("Content-Type", "application/json")
 		w.Write([]byte(`{"id":"msg-pass","type":"message","role":"assistant","model":"claude-3-opus","content":[{"type":"text","text":"passthrough"}],"stop_reason":"end_turn","usage":{"input_tokens":5,"output_tokens":1}}`))
 	})
 
-	createProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-3-opus", upstream.URL)
+	createProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-3-opus", upstream.URL)
 
 	reqBody := map[string]any{
-		"model":      "claude-3-opus",
+		"model":      "anthropic_p/claude-3-opus",  // 客户端发送统一 ID
 		"max_tokens": 1024,
 		"messages":   []map[string]any{{"role": "user", "content": "test"}},
 	}
@@ -321,7 +325,8 @@ func TestConversion_AnthropicToAnthropic_Passthrough(t *testing.T) {
 	r.ServeHTTP(w, req)
 
 	assert.Equal(t, 200, w.Code)
-	assert.Contains(t, w.Body.String(), "passthrough")
+	// Smart Passthrough: 响应体中的上游模型名应被改写为统一 ID
+	assert.Contains(t, w.Body.String(), `"model":"anthropic_p/claude-3-opus"`)
 }
 
 // ============ 流式转换测试 ============
@@ -349,10 +354,10 @@ func TestConversion_OpenAIToAnthropic_Stream(t *testing.T) {
 		}
 	})
 
-	createProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-3-opus", upstream.URL)
+	createProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-3-opus", upstream.URL)
 
 	openaiReq := map[string]any{
-		"model":    "claude-3-opus",
+		"model":    "anthropic_p/claude-3-opus",
 		"messages": []map[string]any{{"role": "user", "content": "Hello"}},
 		"stream":   true,
 	}
@@ -390,10 +395,10 @@ func TestConversion_AnthropicToOpenAI_Stream(t *testing.T) {
 		}
 	})
 
-	createProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	createProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	anthropicReq := map[string]any{
-		"model":      "gpt-4",
+		"model":      "openai_p/gpt-4",
 		"max_tokens": 1024,
 		"messages":   []map[string]any{{"role": "user", "content": "Hello"}},
 		"stream":     true,
@@ -512,7 +517,7 @@ func TestConversion_ProviderWithProtocol(t *testing.T) {
 
 	// 创建带 protocol 字段的 provider
 	providerBody := map[string]any{
-		"id":       "test-protocol",
+		"id":       "test_protocol",
 		"name":     "Test Protocol",
 		"api_key":  "sk-test",
 		"base_url": "https://test.com",
@@ -533,7 +538,7 @@ func TestConversion_ProviderWithProtocol(t *testing.T) {
 
 	// 获取时应包含 protocol
 	w = httptest.NewRecorder()
-	req = httptest.NewRequest("GET", "/api/providers/test-protocol", nil)
+	req = httptest.NewRequest("GET", "/api/providers/test_protocol", nil)
 	r.ServeHTTP(w, req)
 	assert.Equal(t, 200, w.Code)
 
@@ -547,7 +552,7 @@ func TestConversion_ProviderDefaultProtocol(t *testing.T) {
 
 	// 不指定 protocol，默认应为 openai
 	providerBody := map[string]any{
-		"id":       "default-proto",
+		"id":       "default_proto",
 		"name":     "Default",
 		"api_key":  "sk-test",
 		"base_url": "https://test.com",

backend/tests/integration/e2e_conversion_test.go

@@ -8,8 +8,6 @@ import (
 	"io"
 	"net/http"
 	"net/http/httptest"
-	"os"
-	"path/filepath"
 	"strings"
 	"testing"
 	"time"
@@ -17,10 +15,7 @@ import (
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"gorm.io/driver/sqlite"
-	"gorm.io/gorm"
 
-	"nex/backend/internal/config"
 	"nex/backend/internal/conversion"
 	"nex/backend/internal/conversion/anthropic"
 	openaiConv "nex/backend/internal/conversion/openai"
@@ -40,25 +35,20 @@ func setupE2ETest(t *testing.T) (*gin.Engine, *httptest.Server) {
 		w.Write([]byte(`{"error":"not mocked"}`))
 	}))
 
-	dir, _ := os.MkdirTemp("", "e2e-test-*")
+	db := setupTestDB(t)
-	db, err := gorm.Open(sqlite.Open(filepath.Join(dir, "test.db")), &gorm.Config{})
-	require.NoError(t, err)
-	err = db.AutoMigrate(&config.Provider{}, &config.Model{}, &config.UsageStats{})
-	require.NoError(t, err)
 	t.Cleanup(func() {
 		sqlDB, _ := db.DB()
 		if sqlDB != nil {
 			sqlDB.Close()
 		}
 		upstream.Close()
-		os.RemoveAll(dir)
 	})
 
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
 	statsRepo := repository.NewStatsRepository(db)
 
-	providerService := service.NewProviderService(providerRepo)
+	providerService := service.NewProviderService(providerRepo, modelRepo)
 	modelService := service.NewModelService(modelRepo, providerRepo)
 	routingService := service.NewRoutingService(modelRepo, providerRepo)
 	statsService := service.NewStatsService(statsRepo)
@@ -105,7 +95,7 @@ func e2eCreateProviderAndModel(t *testing.T, r *gin.Engine, providerID, protocol
 	require.Equal(t, 201, w.Code)
 
 	modelBody, _ := json.Marshal(map[string]string{
-		"id": modelName, "provider_id": providerID, "model_name": modelName,
+		"provider_id": providerID, "model_name": modelName,
 	})
 	w = httptest.NewRecorder()
 	req = httptest.NewRequest("POST", "/api/models", bytes.NewReader(modelBody))
@@ -178,10 +168,10 @@ func TestE2E_OpenAI_NonStream_BasicText(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{
 			{"role": "user", "content": "你好"},
 		},
@@ -195,7 +185,7 @@ func TestE2E_OpenAI_NonStream_BasicText(t *testing.T) {
 	var resp map[string]any
 	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
 	assert.Equal(t, "chat.completion", resp["object"])
-	assert.Equal(t, "gpt-4o", resp["model"])
+	assert.Equal(t, "openai_p/gpt-4o", resp["model"])
 
 	choices := resp["choices"].([]any)
 	require.Len(t, choices, 1)
@@ -231,10 +221,10 @@ func TestE2E_OpenAI_NonStream_MultiTurn(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 100, "completion_tokens": 20, "total_tokens": 120},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{
 			{"role": "system", "content": "你是编程助手"},
 			{"role": "user", "content": "什么是interface?"},
@@ -279,10 +269,10 @@ func TestE2E_OpenAI_NonStream_ToolCalls(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 80, "completion_tokens": 18, "total_tokens": 98},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{
 			{"role": "user", "content": "北京天气"},
 		},
@@ -335,10 +325,10 @@ func TestE2E_OpenAI_NonStream_MaxTokens_Length(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 20, "completion_tokens": 30, "total_tokens": 50},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "介绍AI历史"}},
 		"max_tokens": 30,
 	})
@@ -372,10 +362,10 @@ func TestE2E_OpenAI_NonStream_UsageWithReasoning(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "o3", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "o3", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "o3",
+		"model": "openai_p/o3",
 		"messages": []map[string]any{{"role": "user", "content": "15+23*2=?"}},
 	})
 	w := httptest.NewRecorder()
@@ -413,10 +403,10 @@ func TestE2E_OpenAI_NonStream_Refusal(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 12, "completion_tokens": 35, "total_tokens": 47},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "做坏事"}},
 	})
 	w := httptest.NewRecorder()
@@ -455,10 +445,10 @@ func TestE2E_OpenAI_Stream_Text(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "你好"}},
 		"stream": true,
 	})
@@ -499,10 +489,10 @@ func TestE2E_OpenAI_Stream_ToolCalls(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"type": "function",
@@ -548,10 +538,10 @@ func TestE2E_OpenAI_Stream_WithUsage(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "hi"}},
 		"stream": true,
 	})
@@ -583,10 +573,10 @@ func TestE2E_Anthropic_NonStream_BasicText(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 15, "output_tokens": 25},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "你好"}},
 	})
 	w := httptest.NewRecorder()
@@ -599,7 +589,7 @@ func TestE2E_Anthropic_NonStream_BasicText(t *testing.T) {
 	require.NoError(t, json.Unmarshal(w.Body.Bytes(), &resp))
 	assert.Equal(t, "message", resp["type"])
 	assert.Equal(t, "assistant", resp["role"])
-	assert.Equal(t, "claude-opus-4-7", resp["model"])
+	assert.Equal(t, "anthropic_p/claude-opus-4-7", resp["model"])
 	assert.Equal(t, "end_turn", resp["stop_reason"])
 
 	content := resp["content"].([]any)
@@ -629,10 +619,10 @@ func TestE2E_Anthropic_NonStream_WithSystem(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 30, "output_tokens": 15},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"system": "你是编程助手",
 		"messages": []map[string]any{{"role": "user", "content": "什么是递归?"}},
 	})
@@ -658,10 +648,10 @@ func TestE2E_Anthropic_NonStream_ToolUse(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 180, "output_tokens": 42},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"name": "get_weather", "description": "获取天气",
@@ -704,10 +694,10 @@ func TestE2E_Anthropic_NonStream_Thinking(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 95, "output_tokens": 280},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 4096,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 4096,
 		"messages": []map[string]any{{"role": "user", "content": "15+23*2=?"}},
 		"thinking": map[string]any{"type": "enabled", "budget_tokens": 2048},
 	})
@@ -736,10 +726,10 @@ func TestE2E_Anthropic_NonStream_MaxTokens(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 22, "output_tokens": 20},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 20,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 20,
 		"messages": []map[string]any{{"role": "user", "content": "介绍AI历史"}},
 	})
 	w := httptest.NewRecorder()
@@ -764,10 +754,10 @@ func TestE2E_Anthropic_NonStream_StopSequence(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 22, "output_tokens": 10},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "从1数到10"}},
 		"stop_sequences": []string{"5"},
 	})
@@ -800,10 +790,10 @@ func TestE2E_Anthropic_NonStream_MetadataUserID(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 12, "output_tokens": 5},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "你好"}},
 		"metadata": map[string]any{"user_id": "user_12345"},
 	})
@@ -829,10 +819,10 @@ func TestE2E_Anthropic_NonStream_UsageWithCache(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"system": []map[string]any{{"type": "text", "text": "你是编程助手。"}},
 		"messages": []map[string]any{{"role": "user", "content": "你好"}},
 	})
@@ -874,10 +864,10 @@ func TestE2E_Anthropic_Stream_Text(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "你好"}},
 		"stream": true,
 	})
@@ -921,10 +911,10 @@ func TestE2E_Anthropic_Stream_Thinking(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 4096,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 4096,
 		"messages": []map[string]any{{"role": "user", "content": "1+1=?"}},
 		"thinking": map[string]any{"type": "enabled", "budget_tokens": 1024},
 		"stream": true,
@@ -970,10 +960,10 @@ func TestE2E_CrossProtocol_OpenAIToAnthropic_RequestFormat(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 10, "output_tokens": 5},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-model", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-model", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-model",
+		"model": "anthropic_p/claude-model",
 		"messages": []map[string]any{{"role": "user", "content": "Hello"}},
 	})
 	w := httptest.NewRecorder()
@@ -1011,10 +1001,10 @@ func TestE2E_CrossProtocol_AnthropicToOpenAI_RequestFormat(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 10, "completion_tokens": 8, "total_tokens": 18},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4", "max_tokens": 1024,
+		"model": "openai_p/gpt-4", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "Hello"}},
 	})
 	w := httptest.NewRecorder()
@@ -1052,10 +1042,10 @@ func TestE2E_CrossProtocol_OpenAIToAnthropic_Stream(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-model", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-model", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-model",
+		"model": "anthropic_p/claude-model",
 		"messages": []map[string]any{{"role": "user", "content": "Hello"}},
 		"stream": true,
 	})
@@ -1092,10 +1082,10 @@ func TestE2E_CrossProtocol_AnthropicToOpenAI_Stream(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4", "max_tokens": 1024,
+		"model": "openai_p/gpt-4", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "Hello"}},
 		"stream": true,
 	})
@@ -1130,10 +1120,10 @@ func TestE2E_OpenAI_ErrorResponse(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "nonexistent", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "nonexistent", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "nonexistent",
+		"model": "openai_p/nonexistent",
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 	})
 	w := httptest.NewRecorder()
@@ -1157,10 +1147,10 @@ func TestE2E_Anthropic_ErrorResponse(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 	})
 	w := httptest.NewRecorder()
@@ -1203,10 +1193,10 @@ func TestE2E_OpenAI_NonStream_ParallelToolCalls(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 100, "completion_tokens": 36, "total_tokens": 136},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "北京和上海的天气"}},
 		"tools": []map[string]any{{
 			"type": "function",
@@ -1255,10 +1245,10 @@ func TestE2E_OpenAI_NonStream_StopSequence(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 10, "completion_tokens": 8, "total_tokens": 18},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "从1数到10"}},
 		"stop":   []string{"5"},
 	})
@@ -1293,10 +1283,10 @@ func TestE2E_OpenAI_NonStream_ContentFilter(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 8, "completion_tokens": 0, "total_tokens": 8},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "危险内容"}},
 	})
 	w := httptest.NewRecorder()
@@ -1325,10 +1315,10 @@ func TestE2E_Anthropic_NonStream_MultiToolUse(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 200, "output_tokens": 84},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "北京和上海的天气"}},
 		"tools": []map[string]any{{
 			"name": "get_weather", "description": "获取天气",
@@ -1374,10 +1364,10 @@ func TestE2E_Anthropic_NonStream_ToolChoiceAny(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 100, "output_tokens": 30},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "现在几点了?"}},
 		"tools": []map[string]any{{
 			"name": "get_time", "description": "获取当前时间",
@@ -1417,10 +1407,10 @@ func TestE2E_Anthropic_NonStream_ArraySystemPrompt(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 50, "output_tokens": 10},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"system": []map[string]any{
 			{"type": "text", "text": "你是编程助手。"},
 			{"type": "text", "text": "请用中文回答。"},
@@ -1454,10 +1444,10 @@ func TestE2E_Anthropic_NonStream_ToolResultMessage(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 150, "output_tokens": 20},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{
 			{"role": "user", "content": "北京天气"},
 			{"role": "assistant", "content": []map[string]any{
@@ -1507,10 +1497,10 @@ func TestE2E_Anthropic_Stream_ToolCalls(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"name": "get_weather", "description": "获取天气",
@@ -1561,10 +1551,10 @@ func TestE2E_CrossProtocol_OpenAIToAnthropic_NonStream_ToolCalls(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 100, "output_tokens": 30},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-model", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-model", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-model",
+		"model": "anthropic_p/claude-model",
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"type": "function",
@@ -1613,10 +1603,10 @@ func TestE2E_CrossProtocol_AnthropicToOpenAI_NonStream_Thinking(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 50, "completion_tokens": 100, "total_tokens": 150},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4", "max_tokens": 4096,
+		"model": "openai_p/gpt-4", "max_tokens": 4096,
 		"messages": []map[string]any{{"role": "user", "content": "宇宙的答案"}},
 	})
 	w := httptest.NewRecorder()
@@ -1643,10 +1633,10 @@ func TestE2E_CrossProtocol_StopReasonMapping(t *testing.T) {
 			"usage": map[string]any{"input_tokens": 10, "output_tokens": 20},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-model", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-model", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-model",
+		"model": "anthropic_p/claude-model",
 		"messages": []map[string]any{{"role": "user", "content": "长文"}},
 	})
 	w := httptest.NewRecorder()
@@ -1685,10 +1675,10 @@ func TestE2E_OpenAI_NonStream_AssistantWithToolResult(t *testing.T) {
 			"usage": map[string]any{"prompt_tokens": 100, "completion_tokens": 20, "total_tokens": 120},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{
 			{"role": "user", "content": "北京天气"},
 			{"role": "assistant", "content": nil, "tool_calls": []map[string]any{{
@@ -1732,10 +1722,10 @@ func TestE2E_CrossProtocol_AnthropicToOpenAI_Stream_ToolCalls(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-model", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-model", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-model",
+		"model": "anthropic_p/claude-model",
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"type": "function",
@@ -1781,10 +1771,10 @@ func TestE2E_CrossProtocol_OpenAIToAnthropic_Stream_ToolCalls(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4", "max_tokens": 1024,
+		"model": "openai_p/gpt-4", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "北京天气"}},
 		"tools": []map[string]any{{
 			"name": "get_weather", "description": "获取天气",
@@ -1819,10 +1809,10 @@ func TestE2E_OpenAI_Upstream5xx_ErrorPassthrough(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "openai-p", "openai", "gpt-4o", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "openai_p", "openai", "gpt-4o", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "gpt-4o",
+		"model": "openai_p/gpt-4o",
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 	})
 	w := httptest.NewRecorder()
@@ -1851,10 +1841,10 @@ func TestE2E_Anthropic_Upstream5xx_ErrorPassthrough(t *testing.T) {
 			},
 		})
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 	})
 	w := httptest.NewRecorder()
@@ -1889,10 +1879,10 @@ func TestE2E_Anthropic_Stream_TruncatedSSE(t *testing.T) {
 			time.Sleep(10 * time.Millisecond)
 		}
 	})
-	e2eCreateProviderAndModel(t, r, "anthropic-p", "anthropic", "claude-opus-4-7", upstream.URL)
+	e2eCreateProviderAndModel(t, r, "anthropic_p", "anthropic", "claude-opus-4-7", upstream.URL)
 
 	body, _ := json.Marshal(map[string]any{
-		"model": "claude-opus-4-7", "max_tokens": 1024,
+		"model": "anthropic_p/claude-opus-4-7", "max_tokens": 1024,
 		"messages": []map[string]any{{"role": "user", "content": "test"}},
 		"stream": true,
 	})

backend/tests/integration/integration_test.go

@@ -9,11 +9,8 @@ import (
 
 	"github.com/gin-gonic/gin"
 	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-	"gorm.io/driver/sqlite"
 	"gorm.io/gorm"
 
-	"nex/backend/internal/config"
 	"nex/backend/internal/domain"
 	"nex/backend/internal/handler"
 	"nex/backend/internal/handler/middleware"
@@ -27,23 +24,13 @@ func init() {
 
 func setupIntegrationTest(t *testing.T) (*gin.Engine, *gorm.DB) {
 	t.Helper()
-	dir := t.TempDir()
+	db := setupTestDB(t)
-	db, err := gorm.Open(sqlite.Open(dir+"/test.db"), &gorm.Config{})
-	require.NoError(t, err)
-	err = db.AutoMigrate(&config.Provider{}, &config.Model{}, &config.UsageStats{})
-	require.NoError(t, err)
-	t.Cleanup(func() {
-		sqlDB, _ := db.DB()
-		if sqlDB != nil {
-			sqlDB.Close()
-		}
-	})
 
 	providerRepo := repository.NewProviderRepository(db)
 	modelRepo := repository.NewModelRepository(db)
 	statsRepo := repository.NewStatsRepository(db)
 
-	providerService := service.NewProviderService(providerRepo)
+	providerService := service.NewProviderService(providerRepo, modelRepo)
 	modelService := service.NewModelService(modelRepo, providerRepo)
 	_ = service.NewRoutingService(modelRepo, providerRepo)
 	statsService := service.NewStatsService(statsRepo)
@@ -97,13 +84,16 @@ func TestOpenAI_CompleteFlow(t *testing.T) {
 
 	// 2. 创建 Model
 	modelBody, _ := json.Marshal(map[string]string{
-		"id": "gpt4", "provider_id": "openai", "model_name": "gpt-4",
+		"provider_id": "openai", "model_name": "gpt-4",
 	})
 	w = httptest.NewRecorder()
 	req = httptest.NewRequest("POST", "/api/models", bytes.NewReader(modelBody))
 	req.Header.Set("Content-Type", "application/json")
 	r.ServeHTTP(w, req)
 	assert.Equal(t, 201, w.Code)
+	var createdModel domain.Model
+	json.Unmarshal(w.Body.Bytes(), &createdModel)
+	assert.NotEmpty(t, createdModel.ID)
 
 	// 3. 列出 Provider
 	w = httptest.NewRecorder()
@@ -135,7 +125,7 @@ func TestOpenAI_CompleteFlow(t *testing.T) {
 
 	// 6. 删除 Model
 	w = httptest.NewRecorder()
-	req = httptest.NewRequest("DELETE", "/api/models/gpt4", nil)
+	req = httptest.NewRequest("DELETE", "/api/models/"+createdModel.ID, nil)
 	r.ServeHTTP(w, req)
 	assert.Equal(t, 204, w.Code)
 
@@ -160,17 +150,19 @@ func TestAnthropic_ModelCreation(t *testing.T) {
 	assert.Equal(t, 201, w.Code)
 
 	modelBody, _ := json.Marshal(map[string]string{
-		"id": "claude3", "provider_id": "anthropic", "model_name": "claude-3-opus-20240229",
+		"provider_id": "anthropic", "model_name": "claude-3-opus-20240229",
 	})
 	w = httptest.NewRecorder()
 	req = httptest.NewRequest("POST", "/api/models", bytes.NewReader(modelBody))
 	req.Header.Set("Content-Type", "application/json")
 	r.ServeHTTP(w, req)
 	assert.Equal(t, 201, w.Code)
+	var createdModel domain.Model
+	json.Unmarshal(w.Body.Bytes(), &createdModel)
 
 	// 验证创建成功
 	w = httptest.NewRecorder()
-	req = httptest.NewRequest("GET", "/api/models/claude3", nil)
+	req = httptest.NewRequest("GET", "/api/models/"+createdModel.ID, nil)
 	r.ServeHTTP(w, req)
 	assert.Equal(t, 200, w.Code)
 }
@@ -188,7 +180,7 @@ func TestStats_RecordingAndQuery(t *testing.T) {
 	r.ServeHTTP(w, req)
 
 	modelBody, _ := json.Marshal(map[string]string{
-		"id": "m1", "provider_id": "p1", "model_name": "gpt-4",
+		"provider_id": "p1", "model_name": "gpt-4",
 	})
 	w = httptest.NewRecorder()
 	req = httptest.NewRequest("POST", "/api/models", bytes.NewReader(modelBody))

backend/tests/integration/testhelper.go

@@ -0,0 +1,37 @@
+package integration
+
+import (
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+	"gorm.io/driver/sqlite"
+	"gorm.io/gorm"
+
+	"nex/backend/internal/config"
+)
+
+// setupTestDB 创建内存 SQLite 数据库并执行 AutoMigrate。
+// 使用 MaxOpenConns(1) 确保 :memory: 模式不会被连接池丢弃。
+func setupTestDB(t *testing.T) *gorm.DB {
+	t.Helper()
+
+	db, err := gorm.Open(sqlite.Open(":memory:"), &gorm.Config{})
+	require.NoError(t, err)
+
+	sqlDB, err := db.DB()
+	require.NoError(t, err)
+	sqlDB.SetMaxOpenConns(1)
+	sqlDB.SetConnMaxLifetime(0)
+
+	err = db.AutoMigrate(&config.Provider{}, &config.Model{}, &config.UsageStats{})
+	require.NoError(t, err)
+
+	t.Cleanup(func() {
+		// 等待异步 goroutine（如 statsService.Record）完成
+		time.Sleep(50 * time.Millisecond)
+		sqlDB.Close()
+	})
+
+	return db
+}

backend/tests/migration_test.go

@@ -0,0 +1,79 @@
+package tests
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"nex/backend/internal/config"
+)
+
+func TestMigration_ModelsUUIDPrimaryKey(t *testing.T) {
+	db := SetupTestDB(t)
+	defer CleanupTestDB(t, db)
+
+	// 创建供应商
+	_ = CreateTestProvider(t, db, "openai")
+
+	// 创建模型使用 UUID 作为 id
+	model, err := CreateTestModel(t, db, "550e8400-e29b-41d4-a716-446655440000", "openai", "gpt-4")
+	require.NoError(t, err)
+
+	// 通过 UUID 查询
+	var result config.Model
+	require.NoError(t, db.First(&result, "id = ?", model.ID).Error)
+	assert.Equal(t, "gpt-4", result.ModelName)
+}
+
+func TestMigration_UniqueProviderModelConstraint(t *testing.T) {
+	db := SetupTestDB(t)
+	defer CleanupTestDB(t, db)
+
+	// 创建供应商
+	_ = CreateTestProvider(t, db, "openai")
+
+	// 创建第一个模型
+	_, err := CreateTestModel(t, db, "uuid-1", "openai", "gpt-4")
+	require.NoError(t, err)
+
+	// 尝试创建相同 (provider_id, model_name) 的模型应失败
+	_, err = CreateTestModel(t, db, "uuid-2", "openai", "gpt-4")
+	assert.Error(t, err, "UNIQUE(provider_id, model_name) 约束应阻止重复")
+
+	// 不同 provider_id 下相同 model_name 应成功
+	_ = CreateTestProvider(t, db, "anthropic")
+	_, err = CreateTestModel(t, db, "uuid-3", "anthropic", "gpt-4")
+	require.NoError(t, err, "不同 provider_id 下相同 model_name 应允许")
+}
+
+func TestMigration_CascadeDelete(t *testing.T) {
+	db := SetupTestDB(t)
+	defer CleanupTestDB(t, db)
+
+	// 创建供应商和模型
+	provider := CreateTestProvider(t, db, "openai")
+	_, err := CreateTestModel(t, db, "uuid-1", "openai", "gpt-4")
+	require.NoError(t, err)
+
+	// 删除供应商应级联删除模型
+	require.NoError(t, db.Delete(&provider).Error)
+
+	var count int64
+	db.Model(&config.Model{}).Where("provider_id = ?", "openai").Count(&count)
+	assert.Equal(t, int64(0), count, "删除供应商后其模型应被级联删除")
+}
+
+func TestMigration_ModelDefaultEnabled(t *testing.T) {
+	db := SetupTestDB(t)
+	defer CleanupTestDB(t, db)
+
+	_ = CreateTestProvider(t, db, "openai")
+
+	// 创建模型不指定 enabled，应默认为 true
+	_, err := CreateTestModel(t, db, "uuid-1", "openai", "gpt-4")
+	require.NoError(t, err)
+
+	var result config.Model
+	require.NoError(t, db.First(&result, "id = ?", "uuid-1").Error)
+	assert.True(t, result.Enabled, "enabled 字段默认应为 true")
+}

openspec/changes/unified-model-id/.openspec.yaml

@@ -1,2 +0,0 @@
-schema: spec-driven
-created: 2026-04-20

openspec/changes/unified-model-id/design.md

@@ -1,137 +0,0 @@
-## Context
-
-Nex 是一个 AI 网关，屏蔽多个 AI 供应商（OpenAI、Anthropic 等）的差异，提供统一的 API 接口。当前后端直接透传上游供应商的原始模型名称（如 `gpt-4`），通过 `models` 表的 `model_name` 字段路由。`models` 表的 `id` 字段当前语义是用户自定义标识符，与上游模型名 `model_name` 之间没有明确的职责分离。
-
-当前架构：
-- `ProxyHandler` 从请求体中提取 `model` 字段 → `RoutingService.Route(modelName)` 按 `model_name` 查询
-- `GET /v1/models` 直接透传到第一个供应商的上游接口
-- `GET /v1/models/{id}` 直接透传到上游
-- `TargetProvider.ModelName` 在 encoder 中覆盖请求体的 `model` 字段
-
-## Goals / Non-Goals
-
-**Goals:**
-- 定义统一模型 ID 格式 `provider_id/model_name`，全局唯一标识一个模型
-- 拦截 `/v1/models` 和 `/v1/models/{unified_id}` 接口，从数据库聚合返回，不再透传上游
-- 所有代理接口（Chat、Embeddings、Rerank）使用统一模型 ID 路由，响应中 `model` 字段覆写为统一 ID
-- `models.id` 改为 UUID（内部标识），`models.model_name` 存储上游供应商的模型名称
-- `provider_id` 约束为 `[a-zA-Z0-9_]+`，防止特殊字符影响 URL 和 JSON 交互
-- 保持协议无关、供应商无关的设计
-
-**Non-Goals:**
-- 不支持供应商别名或模型别名
-- 不做上游模型列表自动同步（管理员手动配置可见模型）
-- 不适配前端（后续统一适配）
-
-## Decisions
-
-### D1: 统一模型 ID 格式 — `provider_id/model_name`
-
-格式: `{provider_id}/{model_name}`，例如 `openai/gpt-4`、`anthropic/claude-3-opus-20240229`。
-
-- 使用 `strings.SplitN(id, "/", 2)` 解析，只在第一个 `/` 处分割
-- `provider_id` 约束为 `[a-zA-Z0-9_]+`，保证不含 `/`，解析安全
-- `model_name`（上游模型名）不受字符约束，因为它不出现在管理 API 的 URL 主键中
-
-选择此格式而非 `provider_id:model_name`（冒号分隔）的原因：斜杠在 JSON 字符串中天然安全，且在 URL 路径中语义清晰（`/v1/models/openai/gpt-4`），更符合 REST 风格。
-
-### D2: models 表 schema 变更
-
-```
-旧: id(TEXT PK, 用户自定义), provider_id, model_name(上游模型名), enabled, created_at
-新: id(UUID PK, 自动生成),   provider_id, model_name(上游模型名), enabled, created_at
-    UNIQUE(provider_id, model_name)
-```
-
-关键语义变化：
-- `id` 从用户自定义标识符变为 UUID 内部主键（自动生成），用于管理接口 CRUD
-- `model_name` 语义不变，始终存储上游供应商的模型名称，发给上游的实际值
-- 新增联合唯一约束 `UNIQUE(provider_id, model_name)` 保证同一供应商内模型不重复
-
-选择保留 `id` 作为 PK 而非使用 `(provider_id, model_name)` 联合主键的原因：上游模型名可能含 `/` 等特殊字符（如 Azure OpenAI 的 deployment 路径），不适合作为管理接口的 URL 参数。`id` 为 UUID 可以避免所有特殊字符问题。
-
-### D3: Models/ModelInfo 接口本地聚合
-
-`GET /v1/models` 从数据库查询所有 `enabled` 的模型（JOIN providers），组装为 `CanonicalModelList`，`ID` 字段使用统一模型 ID，通过客户端协议的 adapter 编码返回。不请求上游。
-
-`GET /v1/models/{provider_id}/{model_name}` 从 URL 提取统一模型 ID，解析后查询数据库，组装为 `CanonicalModelInfo` 返回。不请求上游。
-
-选择纯 DB 聚合而非实时查询上游的原因：
-1. 管理员通过 `/api/models` 控制哪些模型对用户可见，网关的意义在于控制可见性
-2. 响应速度快，不依赖上游可用性
-3. 符合当前架构中管理员手动配置 provider 和 model 的设计哲学
-
-### D4: 跨协议响应 model 字段覆写
-
-跨协议场景下，上游返回的响应经过 decode → encode 全量转换。上游响应中的 `model` 字段是原生模型名（如 `gpt-4`），需要在返回给客户端前覆写为统一模型 ID。
-
-实现位置：`ConversionEngine.ConvertHttpResponse` 新增 `modelOverride string` 参数。在解码上游响应到 canonical 后、编码客户端响应前，将 `canonical.Model` 设为 `modelOverride`。流式场景同理，`CreateStreamConverter` 同样接收 `modelOverride` 参数。
-
-此方案仅在跨协议转换路径使用。选择在 canonical 层面处理的原因：
-1. 跨协议必须全量 decode → encode，canonical 的 Model 字段天然可覆写
-2. 不侵入各协议 adapter 的实现
-3. 与 Smart Passthrough 互补——跨协议不可保真，canonical 覆写是自然的
-
-### D5: ProtocolAdapter 接口扩展
-
-在 `ProtocolAdapter` 接口新增四个方法，将所有协议相关的 model 字段知识归属到 adapter：
-
-1. `ExtractUnifiedModelID(nativePath string) (string, error)` — 从路径中提取统一模型 ID
-2. `ExtractModelName(body []byte, ifaceType InterfaceType) (string, error)` — 从请求体中提取 model 值（所有流程复用，替代 handler 层硬编码的 `extractModelName`）
-3. `RewriteRequestModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error)` — 最小化 JSON 改写请求体中的 model 字段（Smart Passthrough 请求方向）
-4. `RewriteResponseModelName(body []byte, newModel string, ifaceType InterfaceType) ([]byte, error)` — 最小化 JSON 改写响应体中的 model 字段（Smart Passthrough 响应方向）
-
-拆分请求/响应方向的原因：请求体和响应体的 JSON 结构可能不同，model 字段的位置可能不同（当前 OpenAI/Anthropic 协议碰巧都在顶层 `"model"`，但未来协议不一定）。拆分后 adapter 各自独立实现，各自按 ifaceType 分派。
-
-`ExtractModelName` 和两个 `Rewrite*` 方法均接收 `InterfaceType` 参数，因为不同接口类型的请求体/响应体结构可能不同，adapter 按 ifaceType 分派具体的定位和改写逻辑。
-
-对于 `isModelInfoPath` 的调整：允许 suffix 中包含 `/`，因为统一模型 ID 格式为 `provider_id/model_name`。
-
-将此方法放在适配器接口而非 handler 中通用实现的原因：不同协议的模型详情路径格式和请求体结构可能不同，各自拥有独立演进能力。
-
-### D6: provider_id 字符集约束
-
-创建供应商时校验 `id` 字段必须匹配 `^[a-zA-Z0-9_]+$`，长度 1-64。
-
-选择严格限制而非仅排除 `/` 的原因：统一模型 ID 出现在 URL 路径和 JSON 中，`?`、`#`、`&`、`=` 等字符会在 URL 中引起解析问题。限制为字母数字下划线后，URL 中永远安全，不需要编码。
-
-### D7: pkg/modelid 工具包
-
-新增 `pkg/modelid` 包，提供：
-- `ParseUnifiedModelID(id string) (providerID, modelName string, error)` — 解析
-- `FormatUnifiedModelID(providerID, modelName string) string` — 格式化
-- `ValidateProviderID(id string) error` — 校验供应商 ID
-- `IsValidUnifiedModelID(id string) bool` — 校验统一模型 ID
-
-使用标准库 `strings.SplitN` 和 `regexp` 实现，不引入新依赖。
-
-### D8: 同协议 Smart Passthrough
-
-当前同协议透传将请求体原样转发，跳过 decode → encode，保持参数完全保真。但统一模型 ID 要求改写 model 字段，原样透传无法满足。
-
-**Smart Passthrough**：保留同协议透传的保真优势，通过 `json.RawMessage` 做最小化改写。
-
-实现方式：adapter 的 `RewriteRequestModelName` 和 `RewriteResponseModelName` 方法各自解析 JSON 为 `map[string]json.RawMessage`，只替换 model 字段的 value，其余字段保留原始 bytes，不经过任何类型转换。参数保真、不丢精度、不改字段顺序。
-
-各接口类型策略：
-- Chat/Embedding/Rerank（同协议）：Smart Passthrough — 请求改写 model（统一 ID → 上游名），响应改写 model（上游名 → 统一 ID）
-- Chat/Embedding/Rerank（跨协议）：全量 decode → encode + modelOverride
-- Models/ModelInfo：本地数据库聚合，不请求上游
-- Passthrough（未知路径）：原样透传，不改写 model
-
-选择让 adapter 拥有完整协议知识（而非通用 json hack）的原因：
-1. 不同协议的 model 字段位置可能不同，adapter 按 InterfaceType 分派
-2. 请求和响应的 model 字段位置可能不同，拆分 RewriteRequestModelName/RewriteResponseModelName 各自独立实现
-3. adapter 内部实现 `ExtractModelName` 和两个 `Rewrite*` 方法可共享同一份"model 在哪"的定位逻辑
-4. 所有流程复用 `ExtractModelName`，同协议额外复用 `RewriteRequestModelName` + `RewriteResponseModelName`
-
-## Risks / Trade-offs
-
-- **[BREAKING CHANGE]** 代理接口 model 字段格式变更，现有客户端必须适配 → 统一 ID 格式简单直观，服务尚未上线无旧客户端
-- **[联合唯一约束]** 同一供应商下相同 model_name 不允许重复 → 这是正确的行为，语义上就不应该重复
-- **[model_name 含特殊字符]** 上游模型名可能含 `/`（如 Azure deployment 路径）→ 解析用 `SplitN("/", 2)` 安全，管理接口用 `id` 定位不受影响，代理接口中统一 ID 出现在 JSON body 和 URL 路径中均安全
-- **[流式响应覆写]** 同协议流式场景需逐 SSE chunk 调用 RewriteResponseModelName → 每个 chunk 多一次轻量 JSON 解析，用 json.RawMessage 保证开销极小
-
-## Open Questions
-
-无。所有关键决策已在探索阶段确认。

openspec/changes/unified-model-id/proposal.md

@@ -1,41 +0,0 @@
-## Why
-
-当前网关直接透传上游供应商的原始模型名称（如 `gpt-4`、`claude-3-opus`），无法在多供应商场景下唯一标识一个模型。不同供应商可能存在同名模型，客户端无法区分应路由到哪个供应商。网关作为屏蔽供应商差异的统一入口，需要定义自有的模型标识体系，让客户端通过统一的 model ID 访问任意供应商的模型，同时拦截 `/v1/models` 等模型查询接口，聚合所有供应商的模型信息返回。
-
-## What Changes
-
-- **BREAKING**: 引入统一模型 ID 格式 `provider_id/model_name`（如 `openai/gpt-4`），所有代理接口（Chat、Embeddings、Rerank）的 `model` 字段必须使用此格式
-- **BREAKING**: `models` 表主键 `id` 改为 UUID 自动生成（不再由用户提供），`model_name` 字段语义保持不变（存储上游供应商模型名称），新增 `UNIQUE(provider_id, model_name)` 联合唯一约束
-- **BREAKING**: `provider_id` 限制为 `[a-zA-Z0-9_]+` 字符集，禁止特殊字符
-- `GET /v1/models` 改为从数据库聚合返回所有已启用模型，不再透传到上游供应商
-- `GET /v1/models/{unified_id}` 改为从数据库查询返回模型详情，不再透传到上游供应商
-- 同协议透传改为 Smart Passthrough：通过 `json.RawMessage` 最小化改写 model 字段，保持其余参数完全保真
-- 跨协议转换路径：通过 canonical 层面 modelOverride 参数覆写响应 model 字段
-- 管理 API (`/api/models`) 请求体字段适配，响应中新增 `unified_id` 字段
-- 新增 `pkg/modelid` 工具包，提供统一模型 ID 的解析、格式化、校验
-- ProtocolAdapter 接口新增 `ExtractUnifiedModelID`、`ExtractModelName`、`RewriteRequestModelName`、`RewriteResponseModelName` 方法，协议无关地处理 model 字段
-
-## Capabilities
-
-### New Capabilities
-- `unified-model-id`: 统一模型 ID 的解析、格式化、校验工具包，以及 `provider_id` 字符集约束
-
-### Modified Capabilities
-- `model-management`: 模型表结构调整（id 改 UUID 自动生成、新增联合唯一约束），CRUD 接口字段变更（创建不再提供 id）
-- `provider-management`: provider_id 创建时增加字符集校验（`[a-zA-Z0-9_]+`）
-- `unified-proxy-handler`: 统一模型 ID 解析路由、Models/ModelInfo 接口改为本地聚合、同协议 Smart Passthrough、跨协议 modelOverride 覆写
-- `conversion-engine`: 跨协议场景下 ConvertHttpResponse 支持 model 覆写参数
-- `protocol-adapter-openai`: isModelInfoPath 适配含 `/` 路径、新增 ExtractUnifiedModelID、ExtractModelName、RewriteRequestModelName、RewriteResponseModelName
-- `protocol-adapter-anthropic`: isModelInfoPath 适配含 `/` 路径、新增 ExtractUnifiedModelID、ExtractModelName、RewriteRequestModelName、RewriteResponseModelName
-- `request-validation`: provider_id 字符集校验规则、模型创建校验适配
-- `database-migration`: models 表 schema 变更迁移（DROP + CREATE 重建）
-- `usage-statistics`: 明确统计记录使用 providerID + modelName 的上游模型名
-
-## Impact
-
-- **数据库**: models 表 schema 变更（DROP + CREATE 重建）
-- **API 兼容性**: 代理接口 model 字段格式为 BREAKING CHANGE，需客户端适配
-- **管理 API**: `/api/models` 请求体变更（创建不再提供 id，自动生成 UUID），响应新增 unified_id 字段
-- **代码模块**: domain、repository、service、handler、conversion、adapter 层均有改动
-- **测试**: routing service、proxy handler、adapter、model handler 需要新增/更新测试
-- **前端**: 本次变更不涉及前端适配，前端后续统一适配

openspec/changes/unified-model-id/specs/conversion-engine/spec.md

@@ -1,49 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 跨协议响应转换支持 model 覆写
-
-ConversionEngine SHALL 在跨协议响应转换时支持 model 字段覆写。
-
-#### Scenario: ConvertHttpResponse 接收 modelOverride 参数
-
-- **WHEN** 调用 `ConvertHttpResponse` 时传入 `modelOverride` 参数（跨协议场景，非空字符串）
-- **THEN** SHALL 在解码上游响应到 canonical 后，将 `Model` 字段设为 `modelOverride`
-- **THEN** SHALL 使用覆写后的 canonical 编码为客户端协议格式
-
-#### Scenario: modelOverride 为空
-
-- **WHEN** 调用 `ConvertHttpResponse` 时 `modelOverride` 为空字符串
-- **THEN** SHALL NOT 覆写 canonical 的 Model 字段，保持上游原始值
-
-#### Scenario: Chat 响应 model 覆写
-
-- **WHEN** 跨协议转换 Chat 类型响应且 `modelOverride` 非空
-- **THEN** `CanonicalResponse.Model` SHALL 被设为 `modelOverride`
-
-#### Scenario: Embedding 响应 model 覆写
-
-- **WHEN** 跨协议转换 Embedding 类型响应且 `modelOverride` 非空
-- **THEN** `CanonicalEmbeddingResponse.Model` SHALL 被设为 `modelOverride`
-
-#### Scenario: Rerank 响应 model 覆写
-
-- **WHEN** 跨协议转换 Rerank 类型响应且 `modelOverride` 非空
-- **THEN** `CanonicalRerankResponse.Model` SHALL 被设为 `modelOverride`
-
-### Requirement: 跨协议流式转换支持 model 覆写
-
-ConversionEngine SHALL 在跨协议流式转换时支持 model 字段覆写。
-
-#### Scenario: CreateStreamConverter 接收 modelOverride 参数
-
-- **WHEN** 调用 `CreateStreamConverter` 时传入 `modelOverride` 参数（跨协议场景）
-- **THEN** SHALL 在流式 canonical 事件中将 `Model` 字段设为 `modelOverride`
-
-### Requirement: TargetProvider 字段语义
-
-TargetProvider 的 ModelName 字段 SHALL 存储上游供应商的模型名称（即 `model_name` 字段值），语义保持不变。
-
-#### Scenario: encoder 使用 TargetProvider.ModelName
-
-- **WHEN** 协议适配器编码请求时
-- **THEN** SHALL 使用 `TargetProvider.ModelName` 作为发给上游的 `model` 字段值（值为路由结果中的 model_name）

openspec/changes/unified-model-id/specs/database-migration/spec.md

@@ -1,13 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: models 表 schema 变更
-
-系统 SHALL 通过迁移脚本重建 models 表结构（服务未上线，无需考虑数据迁移）。
-
-#### Scenario: 迁移后 models 表结构
-
-- **WHEN** 执行迁移
-- **THEN** SHALL 先 DROP 已有的 models 表（无旧数据）
-- **THEN** SHALL CREATE 新的 models 表，包含字段：id（TEXT PRIMARY KEY）、provider_id（TEXT NOT NULL）、model_name（TEXT NOT NULL）、enabled（INTEGER DEFAULT 1）、created_at（DATETIME）
-- **THEN** SHALL 存在 UNIQUE(provider_id, model_name) 约束
-- **THEN** SHALL 存在 FOREIGN KEY (provider_id) REFERENCES providers(id) ON DELETE CASCADE

openspec/changes/unified-model-id/specs/model-management/spec.md

@@ -1,105 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 创建模型配置
-
-网关 SHALL 允许为供应商创建新的模型配置。
-
-#### Scenario: 使用有效数据创建模型
-
-- **WHEN** 向 `/api/models` 发送 POST 请求，携带有效的模型数据（provider_id, model_name），不提供 id 字段
-- **THEN** 网关 SHALL 自动生成 UUID 作为模型 id
-- **THEN** 网关 SHALL 在数据库中创建新的模型记录
-- **THEN** 网关 SHALL 返回创建的模型，状态码为 201
-- **THEN** 模型 SHALL 默认启用
-- **THEN** 返回的模型 SHALL 包含 `unified_id` 字段，值为 `{provider_id}/{model_name}`
-
-#### Scenario: 使用不存在的供应商创建模型
-
-- **WHEN** 向 `/api/models` 发送 POST 请求，携带不存在的 provider_id
-- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
-- **THEN** 错误 SHALL 指示供应商不存在
-
-#### Scenario: 创建重复模型
-
-- **WHEN** 向 `/api/models` 发送 POST 请求，携带已存在的 provider_id + model_name 组合
-- **THEN** 网关 SHALL 返回错误，状态码为 409 (Conflict)
-- **THEN** 错误 SHALL 指示该供应商下已存在相同模型
-
-### Requirement: 列出所有模型
-
-网关 SHALL 允许获取所有模型配置。
-
-#### Scenario: 成功列出模型
-
-- **WHEN** 向 `/api/models` 发送 GET 请求
-- **THEN** 网关 SHALL 返回所有模型的列表
-- **THEN** 每个模型 SHALL 包含 id, provider_id, model_name, unified_id, enabled, created_at
-
-**变更说明：** 响应新增 unified_id 字段，移除旧语义的 id 自定义输入。
-
-### Requirement: 按供应商列出模型
-
-网关 SHALL 允许获取特定供应商的模型。
-
-#### Scenario: 列出存在供应商的模型
-
-- **WHEN** 向 `/api/models?provider_id=<provider_id>` 发送 GET 请求
-- **THEN** 网关 SHALL 返回指定供应商的模型列表
-- **THEN** 每个模型 SHALL 包含 unified_id 字段
-
-### Requirement: 更新模型配置
-
-网关 SHALL 允许更新现有模型配置。
-
-#### Scenario: 使用有效数据更新模型
-
-- **WHEN** 向 `/api/models/:id` 发送 PUT 请求，携带有效的模型数据
-- **THEN** 网关 SHALL 更新数据库中的模型记录
-- **THEN** 网关 SHALL 返回更新后的模型
-- **THEN** 返回的模型 SHALL 包含更新后的 unified_id
-
-#### Scenario: 更新模型为重复组合
-
-- **WHEN** 向 `/api/models/:id` 发送 PUT 请求，更新 provider_id 或 model_name 导致与已有记录重复
-- **THEN** 网关 SHALL 返回错误，状态码为 409 (Conflict)
-
-### Requirement: 删除模型配置
-
-网关 SHALL 允许删除模型配置。
-
-#### Scenario: 删除存在的模型
-
-- **WHEN** 向 `/api/models/:id` 发送 DELETE 请求，携带有效的模型 ID
-- **THEN** 网关 SHALL 删除模型记录
-- **THEN** 网关 SHALL 返回状态码 204 (No Content)
-
-### Requirement: 使用 service 层处理业务逻辑
-
-Handler SHALL 通过 ModelService 处理业务逻辑。
-
-#### Scenario: 调用 service 方法
-
-- **WHEN** handler 收到请求
-- **THEN** SHALL 调用对应的 ModelService 方法（Create、Get、List、Update、Delete）
-- **THEN** SHALL 使用 domain.Model 类型
-- **THEN** Create 时 SHALL 调用 `uuid.New()` 生成 id
-
-#### Scenario: 供应商验证和唯一性校验
-
-- **WHEN** 创建或更新模型
-- **THEN** SHALL 在 service 层验证供应商存在
-- **THEN** SHALL 在 service 层验证 provider_id + model_name 联合唯一
-
-### Requirement: 使用 repository 层访问数据
-
-Service SHALL 通过 ModelRepository 访问数据。
-
-#### Scenario: 联合查询
-
-- **WHEN** service 需要按 provider 和 model_name 查询模型
-- **THEN** SHALL 调用 `FindByProviderAndModelName(providerID, modelName)` 方法
-
-#### Scenario: 查询所有启用模型
-
-- **WHEN** proxy handler 需要聚合模型列表
-- **THEN** SHALL 调用 `ListEnabled()` 方法，返回所有 enabled 的模型（关联 enabled 的供应商）

openspec/changes/unified-model-id/specs/protocol-adapter-anthropic/spec.md

@@ -1,71 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 模型详情路径识别
-
-Anthropic 适配器 SHALL 正确识别包含统一模型 ID 的模型详情路径。
-
-#### Scenario: 含斜杠的统一模型 ID 路径
-
-- **WHEN** 路径为 `/v1/models/anthropic/claude-3-opus`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
-
-#### Scenario: 含多段斜杠的统一模型 ID 路径
-
-- **WHEN** 路径为 `/v1/models/azure/deployments/gpt-4`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
-
-#### Scenario: 模型列表路径不受影响
-
-- **WHEN** 路径为 `/v1/models`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModels`
-
-### Requirement: 提取统一模型 ID
-
-Anthropic 适配器 SHALL 从路径中提取统一模型 ID。
-
-#### Scenario: 标准路径提取
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/anthropic/claude-3-opus")`
-- **THEN** SHALL 返回 `"anthropic/claude-3-opus"`
-
-#### Scenario: 复杂路径提取
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/azure/accounts/org/models/gpt-4")`
-- **THEN** SHALL 返回 `"azure/accounts/org/models/gpt-4"`
-
-#### Scenario: 非模型详情路径
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models")`
-- **THEN** SHALL 返回错误
-
-### Requirement: 从请求体提取 model
-
-Anthropic 适配器 SHALL 按 InterfaceType 从请求体中提取 model 值。
-
-#### Scenario: Chat 请求提取 model
-
-- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeChat)`，body 为 `{"model":"anthropic/claude-3-opus","messages":[...]}`
-- **THEN** SHALL 返回 `"anthropic/claude-3-opus"`
-
-#### Scenario: 无 model 字段
-
-- **WHEN** 调用 `ExtractModelName(body, ifaceType)`，body 中不含 model 字段
-- **THEN** SHALL 返回空字符串，不返回错误
-
-### Requirement: 最小化改写请求体 model 字段
-
-Anthropic 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写请求体中的 model 字段，其余字段保持原始 bytes。
-
-#### Scenario: 请求体 model 改写（统一 ID → 上游名）
-
-- **WHEN** 调用 `RewriteRequestModelName(body, "claude-3-opus-20240229", InterfaceTypeChat)`
-- **THEN** SHALL 将请求体中 model 字段替换为 `"claude-3-opus-20240229"`，其余字段原样保留
-
-### Requirement: 最小化改写响应体 model 字段
-
-Anthropic 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写响应体中的 model 字段，其余字段保持原始 bytes。请求体和响应体的 model 字段位置可能不同，各自独立实现。
-
-#### Scenario: 响应体 model 改写（上游名 → 统一 ID）
-
-- **WHEN** 调用 `RewriteResponseModelName(body, "anthropic/claude-3-opus", InterfaceTypeChat)`
-- **THEN** SHALL 将响应体中 model 字段替换为 `"anthropic/claude-3-opus"`，其余字段原样保留

openspec/changes/unified-model-id/specs/protocol-adapter-openai/spec.md

@@ -1,89 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 模型详情路径识别
-
-OpenAI 适配器 SHALL 正确识别包含统一模型 ID 的模型详情路径。
-
-#### Scenario: 含斜杠的统一模型 ID 路径
-
-- **WHEN** 路径为 `/v1/models/openai/gpt-4`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
-
-#### Scenario: 含多段斜杠的统一模型 ID 路径
-
-- **WHEN** 路径为 `/v1/models/azure/accounts/org-123/models/gpt-4`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
-
-#### Scenario: 模型列表路径不受影响
-
-- **WHEN** 路径为 `/v1/models`
-- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModels`
-
-### Requirement: 提取统一模型 ID
-
-OpenAI 适配器 SHALL 从路径中提取统一模型 ID。
-
-#### Scenario: 标准路径提取
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/openai/gpt-4")`
-- **THEN** SHALL 返回 `"openai/gpt-4"`
-
-#### Scenario: 复杂路径提取
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/azure/accounts/org/models/gpt-4")`
-- **THEN** SHALL 返回 `"azure/accounts/org/models/gpt-4"`
-
-#### Scenario: 非模型详情路径
-
-- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models")`
-- **THEN** SHALL 返回错误
-
-### Requirement: 从请求体提取 model
-
-OpenAI 适配器 SHALL 按 InterfaceType 从请求体中提取 model 值。
-
-#### Scenario: Chat 请求提取 model
-
-- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeChat)`，body 为 `{"model":"openai/gpt-4","messages":[...]}`
-- **THEN** SHALL 返回 `"openai/gpt-4"`
-
-#### Scenario: Embedding 请求提取 model
-
-- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeEmbeddings)`，body 为 `{"model":"openai/text-embedding-3","input":"text"}`
-- **THEN** SHALL 返回 `"openai/text-embedding-3"`
-
-#### Scenario: 无 model 字段
-
-- **WHEN** 调用 `ExtractModelName(body, ifaceType)`，body 中不含 model 字段
-- **THEN** SHALL 返回空字符串，不返回错误
-
-### Requirement: 最小化改写请求体 model 字段
-
-OpenAI 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写请求体中的 model 字段，其余字段保持原始 bytes。
-
-#### Scenario: 请求体 model 改写（统一 ID → 上游名）
-
-- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeChat)`，body 为 `{"model":"openai/gpt-4","messages":[...],"some_param":"value"}`
-- **THEN** SHALL 返回 `{"model":"gpt-4","messages":[...],"some_param":"value"}`
-- **THEN** 除 model 外的字段 SHALL 保持原始 bytes 不变
-
-#### Scenario: 不同 InterfaceType 的请求改写
-
-- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeEmbeddings)`
-- **THEN** SHALL 按 Embedding 接口的请求体 model 字段位置进行改写
-- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeRerank)`
-- **THEN** SHALL 按 Rerank 接口的请求体 model 字段位置进行改写
-
-### Requirement: 最小化改写响应体 model 字段
-
-OpenAI 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写响应体中的 model 字段，其余字段保持原始 bytes。请求体和响应体的 model 字段位置可能不同，各自独立实现。
-
-#### Scenario: 响应体 model 改写（上游名 → 统一 ID）
-
-- **WHEN** 调用 `RewriteResponseModelName(body, "openai/gpt-4", InterfaceTypeChat)`，body 为上游 Chat 响应
-- **THEN** SHALL 将 model 字段替换为 `"openai/gpt-4"`，其余字段原样保留
-
-#### Scenario: 不同 InterfaceType 的响应改写
-
-- **WHEN** 调用 `RewriteResponseModelName(body, "openai/gpt-4", InterfaceTypeEmbeddings)`
-- **THEN** SHALL 按 Embedding 接口的响应体 model 字段位置进行改写

openspec/changes/unified-model-id/specs/provider-management/spec.md

@@ -1,35 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 创建供应商配置
-
-网关 SHALL 允许通过管理 API 创建新的供应商配置。
-
-#### Scenario: 使用有效数据创建供应商
-
-- **WHEN** 向 `/api/providers` 发送 POST 请求，携带有效的供应商数据（id, name, api_key, base_url, protocol）
-- **THEN** 网关 SHALL 在数据库中创建新的供应商记录
-- **THEN** 网关 SHALL 返回创建的供应商，状态码为 201
-- **THEN** 供应商 SHALL 默认启用
-- **THEN** protocol 字段 SHALL 默认为 "openai"
-
-#### Scenario: 使用重复 ID 创建供应商
-
-- **WHEN** 向 `/api/providers` 发送 POST 请求，携带已存在的 ID
-- **THEN** 网关 SHALL 返回错误，状态码为 409 (Conflict)
-
-#### Scenario: 创建供应商时缺少必需字段
-
-- **WHEN** 向 `/api/providers` 发送 POST 请求，缺少必需字段（id, name, api_key 或 base_url）
-- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
-- **THEN** 错误 SHALL 指示缺少哪些字段
-
-#### Scenario: 创建供应商时 ID 包含非法字符
-
-- **WHEN** 向 `/api/providers` 发送 POST 请求，id 包含非 `[a-zA-Z0-9_]` 字符
-- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
-- **THEN** 错误 SHALL 指示 id 仅允许字母、数字、下划线
-
-#### Scenario: 创建供应商时 ID 过长
-
-- **WHEN** 向 `/api/providers` 发送 POST 请求，id 长度超过 64
-- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)

openspec/changes/unified-model-id/specs/request-validation/spec.md

@@ -1,39 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 供应商 ID 校验
-
-创建供应商时，SHALL 对 `id` 字段进行字符集校验。
-
-#### Scenario: 合法字符集
-
-- **WHEN** 创建供应商，id 仅包含 `[a-zA-Z0-9_]` 字符
-- **THEN** SHALL 校验通过
-
-#### Scenario: 非法字符
-
-- **WHEN** 创建供应商，id 包含 `-`、`.`、`/`、空格、中文等非 `[a-zA-Z0-9_]` 字符
-- **THEN** SHALL 返回 400 错误
-
-#### Scenario: 长度限制
-
-- **WHEN** 创建供应商，id 长度超过 64
-- **THEN** SHALL 返回 400 错误
-
-### Requirement: 模型创建校验
-
-创建模型时，SHALL 对 `provider_id` + `model_name` 进行联合唯一性校验。
-
-#### Scenario: 正常创建
-
-- **WHEN** 创建模型，provider_id 存在且 provider_id + model_name 组合唯一
-- **THEN** SHALL 校验通过
-
-#### Scenario: 联合唯一冲突
-
-- **WHEN** 创建模型，provider_id + model_name 组合已存在
-- **THEN** SHALL 返回 409 错误
-
-#### Scenario: model_name 为空
-
-- **WHEN** 创建模型，未提供 model_name
-- **THEN** SHALL 返回 400 错误

openspec/changes/unified-model-id/specs/unified-proxy-handler/spec.md

@@ -1,123 +0,0 @@
-## MODIFIED Requirements
-
-### Requirement: 代理请求路由
-
-ProxyHandler SHALL 使用统一模型 ID 路由所有代理请求。
-
-#### Scenario: 提取统一模型 ID
-
-- **WHEN** 收到 Chat、Embeddings 或 Rerank 接口的 POST 请求（含请求体）
-- **THEN** SHALL 调用客户端协议 adapter 的 `ExtractModelName(body, ifaceType)` 提取 model 值
-- **THEN** SHALL 调用 `ParseUnifiedModelID` 解析得到 providerID 和 modelName
-- **THEN** SHALL 调用 `RoutingService.RouteByModelName(providerID, modelName)` 路由
-
-#### Scenario: GET 请求或无请求体
-
-- **WHEN** 收到 GET 请求或请求体为空
-- **THEN** SHALL 返回错误响应，状态码为 400，提示缺少 model 字段
-
-#### Scenario: 无效的统一模型 ID
-
-- **WHEN** 请求体中 `model` 字段不是有效的统一模型 ID 格式
-- **THEN** SHALL 返回错误响应，状态码为 400
-
-#### Scenario: 模型不存在
-
-- **WHEN** 解析统一模型 ID 后，数据库中找不到对应的 provider_id + model_name 组合
-- **THEN** SHALL 返回错误响应，状态码为 404
-
-#### Scenario: 模型已禁用
-
-- **WHEN** 解析统一模型 ID 后，对应的模型 enabled 为 false
-- **THEN** SHALL 返回错误响应，状态码为 404
-
-#### Scenario: 供应商已禁用
-
-- **WHEN** 解析统一模型 ID 后，对应的供应商 enabled 为 false
-- **THEN** SHALL 返回错误响应，状态码为 404
-
-### Requirement: 同协议 Smart Passthrough
-
-当客户端协议与供应商协议相同时，ProxyHandler SHALL 使用 Smart Passthrough 处理 Chat、Embedding、Rerank 请求。
-
-#### Scenario: 同协议非流式请求
-
-- **WHEN** 客户端协议 == 供应商协议，且为非流式请求
-- **THEN** SHALL 调用 adapter 的 `RewriteRequestModelName(body, modelName, ifaceType)` 将请求体中 model 从统一 ID 改写为上游模型名
-- **THEN** SHALL 构建 URL 和 Headers（同当前透传逻辑）
-- **THEN** SHALL 发送改写后的请求体到上游
-- **THEN** SHALL 调用 adapter 的 `RewriteResponseModelName(resp.Body, unifiedModelID, ifaceType)` 将响应中 model 从上游名改写为统一 ID
-- **THEN** SHALL NOT 对 body 做全量 decode → encode，保持未改写字段的原始 bytes
-
-#### Scenario: 同协议流式请求
-
-- **WHEN** 客户端协议 == 供应商协议，且为流式请求
-- **THEN** SHALL 对请求体做 `RewriteRequestModelName` 改写 model 字段
-- **THEN** SHALL 逐 SSE chunk 调用 `RewriteResponseModelName` 改写响应中 model 字段
-- **THEN** SHALL NOT 对 chunk 做全量 decode → encode
-
-#### Scenario: Smart Passthrough 保真性
-
-- **WHEN** 客户端发送含未知参数的请求（如 `{"model":"openai/gpt-4","some_new_param":"value"}`）
-- **THEN** 上游 SHALL 收到 `{"model":"gpt-4","some_new_param":"value"}`
-- **THEN** `some_new_param` SHALL 保持原始值不变，不丢失、不改变类型
-
-### Requirement: 跨协议完整转换
-
-当客户端协议与供应商协议不同时，ProxyHandler SHALL 使用全量转换路径。
-
-#### Scenario: 跨协议非流式请求
-
-- **WHEN** 客户端协议 != 供应商协议
-- **THEN** SHALL 走 `ConvertHttpRequest` 全量转换，encoder 中 provider.ModelName 覆盖 model
-- **THEN** SHALL 走 `ConvertHttpResponse` 全量转换，modelOverride 参数覆写 canonical.Model
-
-#### Scenario: 跨协议流式请求
-
-- **WHEN** 客户端协议 != 供应商协议，且为流式请求
-- **THEN** SHALL 走 `CreateStreamConverter` 全量转换，modelOverride 参数覆写流式 canonical 事件中的 Model
-
-### Requirement: 模型列表本地聚合
-
-ProxyHandler SHALL 从数据库聚合返回模型列表，不再透传上游。
-
-#### Scenario: GET /v1/models
-
-- **WHEN** 收到 `GET /{protocol}/v1/models` 请求
-- **THEN** SHALL 从数据库查询所有 enabled 的模型（关联 enabled 的供应商）
-- **THEN** SHALL 组装 `CanonicalModelList`，每个模型的 ID 字段为统一模型 ID（`provider_id/model_name`），Name 字段为 model_name，OwnedBy 字段为 provider_id
-- **THEN** SHALL 使用客户端协议的 adapter 编码响应
-- **THEN** SHALL NOT 请求上游供应商
-
-#### Scenario: 无可用模型
-
-- **WHEN** 数据库中没有 enabled 的模型
-- **THEN** SHALL 返回空列表
-
-### Requirement: 模型详情本地查询
-
-ProxyHandler SHALL 从数据库查询返回模型详情，不再透传上游。
-
-#### Scenario: GET /v1/models/{unified_id}
-
-- **WHEN** 收到 `GET /{protocol}/v1/models/{provider_id}/{model_name}` 请求
-- **THEN** SHALL 调用 adapter 的 `ExtractUnifiedModelID` 提取统一模型 ID
-- **THEN** SHALL 解析统一模型 ID 得到 providerID 和 modelName
-- **THEN** SHALL 从数据库查询对应的模型和供应商
-- **THEN** SHALL 组装 `CanonicalModelInfo`，ID 字段为统一模型 ID（`provider_id/model_name`），Name 字段为 model_name，OwnedBy 字段为 provider_id
-- **THEN** SHALL 使用客户端协议的 adapter 编码响应
-- **THEN** SHALL NOT 请求上游供应商
-
-#### Scenario: 模型详情不存在
-
-- **WHEN** 统一模型 ID 对应的模型不存在或已禁用
-- **THEN** SHALL 返回错误响应，状态码为 404
-
-### Requirement: 统计记录
-
-ProxyHandler SHALL 使用 providerID 和 modelName 记录使用统计。
-
-#### Scenario: 异步记录统计
-
-- **WHEN** 代理请求成功完成
-- **THEN** SHALL 异步调用 `StatsService.Record(providerID, modelName)`

openspec/changes/unified-model-id/specs/usage-statistics/spec.md

@@ -1,16 +0,0 @@
-## ADDED Requirements
-
-### Requirement: 使用统计记录统一模型标识
-
-系统 SHALL 使用 providerID 和 modelName（上游模型名）记录使用统计。
-
-#### Scenario: 代理请求统计记录
-
-- **WHEN** 代理请求成功完成
-- **THEN** SHALL 记录 provider_id 和 model_name 到 usage_stats 表（参数来自路由结果）
-- **THEN** SHALL 异步执行，不阻塞响应
-
-#### Scenario: 查询统计
-
-- **WHEN** 查询统计数据
-- **THEN** 支持按 provider_id 和 model_name 过滤

openspec/changes/unified-model-id/tasks.md

@@ -1,53 +0,0 @@
-## 1. 数据库迁移
-
-- [ ] 1.1 新增迁移脚本：DROP 旧 models 表 + CREATE 新 models 表（id UUID PK, provider_id, model_name, enabled, created_at），UNIQUE(provider_id, model_name)
-- [ ] 1.2 更新 config/models.go：Model 结构体适配（id 改为 UUID 自动生成，model_name 保持不变）
-- [ ] 1.3 编写迁移脚本测试
-
-## 2. 统一模型 ID 工具包
-
-- [ ] 2.1 新增 pkg/modelid/model_id.go：实现 ParseUnifiedModelID、FormatUnifiedModelID、ValidateProviderID、IsValidUnifiedModelID
-- [ ] 2.2 新增 pkg/modelid/model_id_test.go：覆盖标准格式、含斜杠 model_name、空字符串、非法字符等边界情况
-
-## 3. Domain 层适配
-
-- [ ] 3.1 修改 domain/model.go：Model 结构体字段适配，新增 UnifiedModelID() 方法
-- [ ] 3.2 修改 domain/route.go：RouteResult 适配新字段
-
-## 4. Repository 层适配
-
-- [ ] 4.1 修改 repository/model_repo.go：接口变更 — GetByModelName 改为 FindByProviderAndModelName，新增 ListEnabled
-- [ ] 4.2 修改 repository/model_repo_impl.go：实现 FindByProviderAndModelName（WHERE provider_id=? AND model_name=?）、ListEnabled（JOIN providers WHERE enabled）
-- [ ] 4.3 编写 repository 层测试
-
-## 5. Service 层适配
-
-- [ ] 5.1 修改 service/routing_service.go：Route 接口改为 RouteByModelName(providerID, modelName string)
-- [ ] 5.2 修改 service/routing_service_impl.go：调用 FindByProviderAndModelName 替代 GetByModelName
-- [ ] 5.3 修改 service/model_service.go：Create 生成 UUID、新增联合唯一校验方法
-- [ ] 5.4 修改 service/model_service_impl.go：实现联合唯一校验、UUID 生成
-- [ ] 5.5 修改 service/provider_service_impl.go：Create 时调用 ValidateProviderID 校验 ID 字符集
-- [ ] 5.6 编写 service 层测试
-
-## 6. Conversion 层适配
-
-- [ ] 6.1 修改 conversion/adapter.go：ProtocolAdapter 接口新增 ExtractUnifiedModelID、ExtractModelName、RewriteRequestModelName、RewriteResponseModelName 四个方法
-- [ ] 6.2 修改 conversion/engine.go：ConvertHttpResponse 新增 modelOverride 参数（跨协议场景），各 convert*ResponseBody 中覆写 canonical Model；CreateStreamConverter 新增 modelOverride 参数
-- [ ] 6.3 修改 conversion/openai/adapter.go：实现 ExtractUnifiedModelID、ExtractModelName（按 ifaceType 提取 model）、RewriteRequestModelName 和 RewriteResponseModelName（json.RawMessage 最小化改写，按 ifaceType 定位 model 字段，请求/响应独立实现），修改 isModelInfoPath 允许 suffix 含 "/"
-- [ ] 6.4 修改 conversion/anthropic/adapter.go：实现 ExtractUnifiedModelID、ExtractModelName、RewriteRequestModelName、RewriteResponseModelName，修改 isModelInfoPath 允许 suffix 含 "/"
-- [ ] 6.5 编写 conversion 层测试：ExtractUnifiedModelID、ExtractModelName 各 ifaceType、RewriteRequestModelName/RewriteResponseModelName 保真性（含未知参数不丢失）、isModelInfoPath 含斜杠路径、modelOverride 覆写
-
-## 7. Handler 层改造
-
-- [ ] 7.1 修改 handler/proxy_handler.go：HandleProxy 按接口类型分发 — Models/ModelInfo 本地聚合；Chat/Embed/Rerank 用 adapter.ExtractModelName 提取统一 ID 路由，同协议走 Smart Passthrough（adapter.RewriteRequestModelName 改写请求、adapter.RewriteResponseModelName 改写响应），跨协议走全量转换（modelOverride）；删除 forwardPassthrough 和硬编码的 extractModelName
-- [ ] 7.2 修改 handler/model_handler.go：请求体字段适配（移除 id 输入、保留 provider_id 和 model_name），响应新增 unified_id，Create 使用 UUID
-- [ ] 7.3 修改 handler/provider_handler.go：CreateProvider 校验 ID 字符集
-- [ ] 7.4 编写 handler 层测试：统一模型 ID 路由、同协议 Smart Passthrough 保真性、跨协议 modelOverride、Models 聚合、ModelInfo 查询、流式场景 model 覆写、provider ID 校验
-
-## 8. 路由注册适配
-
-- [ ] 8.1 修改 cmd/server/main.go：setupRoutes 适配 handler 签名变更，传递新增依赖
-
-## 9. 文档更新
-
-- [ ] 9.1 按需更新 README.md：同步 models 表结构、API 接口字段、统一模型 ID 格式、Smart Passthrough 策略等变更说明

openspec/specs/conversion-engine/spec.md

@@ -277,4 +277,51 @@ ErrorCode SHALL 包含：INVALID_INPUT、MISSING_REQUIRED_FIELD、INCOMPATIBLE_F
 - **WHEN** Adapter 调用 buildHeaders(provider)
 - **THEN** SHALL 从 provider.api_key 提取认证信息
 - **THEN** SHALL 从 provider.adapter_config 提取协议专属配置
-- **THEN** SHALL 使用 provider.model_name 覆盖请求中的 model 字段
+- **THEN** SHALL 使用 provider.model_name 覆盖请求中的 model 字段
+### Requirement: 跨协议响应转换支持 model 覆写
+
+ConversionEngine SHALL 在跨协议响应转换时支持 model 字段覆写。
+
+#### Scenario: ConvertHttpResponse 接收 modelOverride 参数
+
+- **WHEN** 调用 `ConvertHttpResponse` 时传入 `modelOverride` 参数（跨协议场景，非空字符串）
+- **THEN** SHALL 在解码上游响应到 canonical 后，将 `Model` 字段设为 `modelOverride`
+- **THEN** SHALL 使用覆写后的 canonical 编码为客户端协议格式
+
+#### Scenario: modelOverride 为空
+
+- **WHEN** 调用 `ConvertHttpResponse` 时 `modelOverride` 为空字符串
+- **THEN** SHALL NOT 覆写 canonical 的 Model 字段，保持上游原始值
+
+#### Scenario: Chat 响应 model 覆写
+
+- **WHEN** 跨协议转换 Chat 类型响应且 `modelOverride` 非空
+- **THEN** `CanonicalResponse.Model` SHALL 被设为 `modelOverride`
+
+#### Scenario: Embedding 响应 model 覆写
+
+- **WHEN** 跨协议转换 Embedding 类型响应且 `modelOverride` 非空
+- **THEN** `CanonicalEmbeddingResponse.Model` SHALL 被设为 `modelOverride`
+
+#### Scenario: Rerank 响应 model 覆写
+
+- **WHEN** 跨协议转换 Rerank 类型响应且 `modelOverride` 非空
+- **THEN** `CanonicalRerankResponse.Model` SHALL 被设为 `modelOverride`
+
+### Requirement: 跨协议流式转换支持 model 覆写
+
+ConversionEngine SHALL 在跨协议流式转换时支持 model 字段覆写。
+
+#### Scenario: CreateStreamConverter 接收 modelOverride 参数
+
+- **WHEN** 调用 `CreateStreamConverter` 时传入 `modelOverride` 参数（跨协议场景）
+- **THEN** SHALL 在流式 canonical 事件中将 `Model` 字段设为 `modelOverride`
+
+### Requirement: TargetProvider 字段语义
+
+TargetProvider 的 ModelName 字段 SHALL 存储上游供应商的模型名称（即 `model_name` 字段值），语义保持不变。
+
+#### Scenario: encoder 使用 TargetProvider.ModelName
+
+- **WHEN** 协议适配器编码请求时
+- **THEN** SHALL 使用 `TargetProvider.ModelName` 作为发给上游的 `model` 字段值（值为路由结果中的 model_name）

openspec/specs/database-migration/spec.md

@@ -28,9 +28,10 @@
 #### Scenario: 初始迁移文件
 
 - **WHEN** 创建初始迁移
-- **THEN** SHALL 创建 001_initial_schema.sql
+- **THEN** SHALL 创建单个初始迁移文件（如 `20260421000001_initial_schema.sql`）
 - **THEN** SHALL 包含 providers、models、usage_stats 表的创建语句
 - **THEN** SHALL 包含外键约束
+- **THEN** SHALL 包含索引创建语句
 
 #### Scenario: Up 迁移
 
@@ -42,25 +43,19 @@
 #### Scenario: Down 迁移
 
 - **WHEN** 执行 down 迁移
-- **THEN** SHALL 删除所有表
+- **THEN** SHALL 删除所有表和索引
 - **THEN** SHALL 按正确顺序删除（避免外键约束错误）
 
-### Requirement: 添加索引迁移
+### Requirement: models 表 schema 变更
 
-系统 SHALL 创建索引迁移。
+系统 SHALL 在初始迁移脚本中直接创建新的 models 表结构（服务未上线，无需考虑数据迁移，迁移脚本已合并为单个初始迁移文件）。
 
-#### Scenario: 索引迁移文件
+#### Scenario: 初始迁移 models 表结构
 
-- **WHEN** 创建索引迁移
+- **WHEN** 执行迁移
-- **THEN** SHALL 创建 002_add_indexes.sql
+- **THEN** SHALL CREATE models 表，包含字段：id（TEXT PRIMARY KEY，存储 UUID 字符串）、provider_id（TEXT NOT NULL）、model_name（TEXT NOT NULL）、enabled（INTEGER DEFAULT 1）、created_at（DATETIME）
-- **THEN** SHALL 为常用查询字段添加索引
+- **THEN** SHALL 存在 UNIQUE(provider_id, model_name) 约束
-
+- **THEN** SHALL 存在 FOREIGN KEY (provider_id) REFERENCES providers(id) ON DELETE CASCADE
-#### Scenario: 索引定义
-
-- **WHEN** 添加索引
-- **THEN** SHALL 为 models(provider_id) 添加索引
-- **THEN** SHALL 为 models(model_name) 添加索引
-- **THEN** SHALL 为 usage_stats(provider_id, model_name, date) 添加复合索引
 
 ### Requirement: 迁移命令集成
 

openspec/specs/error-responses/spec.md

@@ -0,0 +1,209 @@
+# Error Responses
+
+## Purpose
+
+定义系统统一的错误响应格式和各类错误场景，确保客户端能够一致地处理错误。
+
+## Requirements
+
+### Requirement: 统一错误响应格式
+
+系统 SHALL 使用统一的错误响应格式。
+
+#### Scenario: 标准错误格式
+
+- **WHEN** 返回错误响应
+- **THEN** SHALL 使用以下 JSON 格式：
+  ```json
+  {
+    "error": "错误描述",
+    "code": "ERROR_CODE"
+  }
+  ```
+- **THEN** `error` 字段 SHALL 包含人类可读的错误描述
+- **THEN** `code` 字段 SHALL 包含机器可读的错误代码（可选）
+
+### Requirement: provider_id 校验错误
+
+系统 SHALL 对 provider_id 校验错误返回明确的错误信息。
+
+#### Scenario: provider_id 包含非法字符
+
+- **WHEN** 创建或更新供应商时，provider_id 包含非 `[a-zA-Z0-9_]` 字符
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "供应商 ID 仅允许字母、数字、下划线",
+    "code": "INVALID_PROVIDER_ID"
+  }
+  ```
+
+#### Scenario: provider_id 长度超限
+
+- **WHEN** 创建或更新供应商时，provider_id 长度超过 64
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "供应商 ID 长度不能超过 64 个字符",
+    "code": "INVALID_PROVIDER_ID"
+  }
+  ```
+
+### Requirement: 联合唯一约束冲突错误
+
+系统 SHALL 对联合唯一约束冲突返回明确的错误信息。
+
+#### Scenario: 创建模型时 provider_id + model_name 组合已存在
+
+- **WHEN** 创建模型时，provider_id + model_name 组合已存在
+- **THEN** SHALL 返回 HTTP 409 Conflict
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "同一供应商下模型名称已存在",
+    "code": "duplicate_model"
+  }
+  ```
+
+#### Scenario: 更新模型时导致 provider_id + model_name 组合冲突
+
+- **WHEN** 更新模型时，修改 provider_id 或 model_name 导致与已有记录冲突
+- **THEN** SHALL 返回 HTTP 409 Conflict
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "同一供应商下模型名称已存在",
+    "code": "duplicate_model"
+  }
+  ```
+
+### Requirement: 资源不存在错误
+
+系统 SHALL 对资源不存在返回明确的错误信息。
+
+#### Scenario: 模型不存在
+
+- **WHEN** 查询或操作不存在的模型
+- **THEN** SHALL 返回 HTTP 404 Not Found
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "模型未找到"
+  }
+  ```
+
+#### Scenario: 供应商不存在
+
+- **WHEN** 创建模型时指定的供应商不存在
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "供应商不存在"
+  }
+  ```
+
+### Requirement: 统一模型 ID 格式错误
+
+系统 SHALL 对统一模型 ID 格式错误返回明确的错误信息。
+
+#### Scenario: 统一模型 ID 格式无效
+
+- **WHEN** 代理请求中的 model 字段不是有效的统一模型 ID 格式
+- **THEN** 请求 SHALL 走 forwardPassthrough 透传到上游（兼容未适配的客户端）
+- **THEN** 不返回错误，保持与上游的兼容性
+
+**设计理由：**
+- 统一模型 ID 是 BREAKING CHANGE，部分旧客户端可能仍使用原始模型名
+- 透传策略允许上游自行判断并返回错误（如 404 model not found）
+- 网关作为透明代理，不应拦截所有格式非法的请求
+
+#### Scenario: 统一模型 ID 格式有效但对应模型不存在
+
+- **WHEN** 代理请求中的 model 字段是有效的统一模型 ID 格式（含 `/`），但数据库中找不到对应的模型
+- **THEN** SHALL 返回 HTTP 404 Not Found
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "模型未找到"
+  }
+  ```
+
+#### Scenario: 统一模型 ID 对应的模型不存在
+
+- **WHEN** 解析统一模型 ID 后，数据库中找不到对应的模型
+- **THEN** SHALL 返回 HTTP 404 Not Found
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "模型未找到"
+  }
+  ```
+
+#### Scenario: 统一模型 ID 对应的模型已禁用
+
+- **WHEN** 解析统一模型 ID 后，对应的模型 enabled 为 false
+- **THEN** SHALL 返回 HTTP 404 Not Found
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "模型未找到"
+  }
+  ```
+
+#### Scenario: 统一模型 ID 对应的供应商已禁用
+
+- **WHEN** 解析统一模型 ID 后，对应的供应商 enabled 为 false
+- **THEN** SHALL 返回 HTTP 404 Not Found
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "模型未找到"
+  }
+  ```
+
+### Requirement: JSON 格式错误
+
+系统 SHALL 对请求体 JSON 格式错误返回明确的错误信息。
+
+#### Scenario: 请求体 JSON 格式错误
+
+- **WHEN** 代理请求的请求体不是有效的 JSON 格式
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "请求体 JSON 格式错误",
+    "code": "INVALID_JSON"
+  }
+  ```
+
+#### Scenario: Smart Passthrough 时请求体 JSON 格式错误
+
+- **WHEN** 同协议 Smart Passthrough 场景下，请求体 JSON 格式不正确
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "请求体 JSON 格式错误",
+    "code": "INVALID_JSON"
+  }
+  ```
+
+### Requirement: 不可变字段错误
+
+系统 SHALL 对尝试修改不可变字段返回明确的错误信息。
+
+#### Scenario: 尝试修改供应商 ID
+
+- **WHEN** 更新供应商时，请求体中包含 `id` 字段
+- **THEN** SHALL 返回 HTTP 400 Bad Request
+- **THEN** SHALL 返回以下 JSON 格式：
+  ```json
+  {
+    "error": "供应商 ID 不允许修改",
+    "code": "IMMUTABLE_FIELD"
+  }
+  ```

openspec/specs/model-management/spec.md

@@ -10,10 +10,12 @@
 
 #### Scenario: 使用有效数据创建模型
 
-- **WHEN** 向 `/api/models` 发送 POST 请求，携带有效的模型数据（id, provider_id, model_name）
+- **WHEN** 向 `/api/models` 发送 POST 请求，携带有效的模型数据（provider_id, model_name），不提供 id 字段
+- **THEN** 网关 SHALL 自动生成 UUID 作为模型 id
 - **THEN** 网关 SHALL 在数据库中创建新的模型记录
 - **THEN** 网关 SHALL 返回创建的模型，状态码为 201
 - **THEN** 模型 SHALL 默认启用
+- **THEN** 返回的模型 SHALL 包含 `unified_id` 字段，值为 `{provider_id}/{model_name}`
 
 #### Scenario: 使用不存在的供应商创建模型
 
@@ -21,7 +23,11 @@
 - **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
 - **THEN** 错误 SHALL 指示供应商不存在
 
-**变更说明：** handler 通过 ModelService 调用，数据访问通过 ModelRepository 和 ProviderRepository。API 接口保持不变。
+#### Scenario: 创建重复模型
+
+- **WHEN** 向 `/api/models` 发送 POST 请求，携带已存在的 provider_id + model_name 组合
+- **THEN** 网关 SHALL 返回错误，状态码为 409 (Conflict)
+- **THEN** 错误 SHALL 指示同一供应商下模型名称已存在
 
 ### Requirement: 列出所有模型
 
@@ -31,9 +37,7 @@
 
 - **WHEN** 向 `/api/models` 发送 GET 请求
 - **THEN** 网关 SHALL 返回所有模型的列表
-- **THEN** 每个模型 SHALL 包含 id, provider_id, model_name, enabled, created_at
+- **THEN** 每个模型 SHALL 包含 id, provider_id, model_name, unified_id, enabled, created_at
-
-**变更说明：** 数据访问从 config 包迁移到 ModelRepository。API 接口保持不变。
 
 ### Requirement: 按供应商列出模型
 
@@ -43,8 +47,7 @@
 
 - **WHEN** 向 `/api/models?provider_id=<provider_id>` 发送 GET 请求
 - **THEN** 网关 SHALL 返回指定供应商的模型列表
-
+- **THEN** 每个模型 SHALL 包含 unified_id 字段
-**变更说明：** 通过 ModelService 和 ModelRepository 实现。API 接口保持不变。
 
 ### Requirement: 更新模型配置
 
@@ -55,14 +58,12 @@
 - **WHEN** 向 `/api/models/:id` 发送 PUT 请求，携带有效的模型数据
 - **THEN** 网关 SHALL 更新数据库中的模型记录
 - **THEN** 网关 SHALL 返回更新后的模型
+- **THEN** 返回的模型 SHALL 包含更新后的 unified_id
 
-#### Scenario: 更新模型供应商
+#### Scenario: 更新模型为重复组合
 
-- **WHEN** 向 `/api/models/:id` 发送 PUT 请求，携带新的 provider_id
+- **WHEN** 向 `/api/models/:id` 发送 PUT 请求，更新 provider_id 或 model_name 导致与已有记录重复
-- **THEN** 网关 SHALL 验证新供应商是否存在
+- **THEN** 网关 SHALL 返回错误，状态码为 409 (Conflict)
-- **THEN** 网关 SHALL 更新模型的供应商关联
-
-**变更说明：** 通过 ModelService、ModelRepository 和 ProviderRepository 实现。API 接口保持不变。
 
 ### Requirement: 删除模型配置
 
@@ -74,8 +75,6 @@
 - **THEN** 网关 SHALL 删除模型记录
 - **THEN** 网关 SHALL 返回状态码 204 (No Content)
 
-**变更说明：** 通过 ModelService 和 ModelRepository 实现。API 接口保持不变。
-
 ### Requirement: 使用 service 层处理业务逻辑
 
 Handler SHALL 通过 ModelService 处理业务逻辑。
@@ -85,25 +84,60 @@ Handler SHALL 通过 ModelService 处理业务逻辑。
 - **WHEN** handler 收到请求
 - **THEN** SHALL 调用对应的 ModelService 方法（Create、Get、List、Update、Delete）
 - **THEN** SHALL 使用 domain.Model 类型
+- **THEN** Create 时 SHALL 调用 `uuid.New()` 生成 id
 
-#### Scenario: 供应商验证
+#### Scenario: 供应商验证和唯一性校验
 
 - **WHEN** 创建或更新模型
 - **THEN** SHALL 在 service 层验证供应商存在
-- **THEN** SHALL 通过 ProviderRepository 查询供应商
+- **THEN** SHALL 在 service 层验证 provider_id + model_name 联合唯一
+
+### Requirement: 联合唯一约束并发处理
+
+创建或更新模型时，SHALL 使用应用层校验 + 数据库约束双重保险处理联合唯一约束。
+
+#### Scenario: 应用层快速失败
+
+- **WHEN** 创建或更新模型前
+- **THEN** SHALL 先检查 provider_id + model_name 是否已存在
+- **THEN** 如已存在，SHALL 返回 HTTP 409 Conflict
+- **THEN** SHALL 返回错误格式：
+  ```json
+  {
+    "error": "同一供应商下模型名称已存在",
+    "code": "duplicate_model"
+  }
+  ```
+
+#### Scenario: 数据库约束兜底
+
+- **WHEN** 并发创建导致应用层校验通过但数据库写入失败
+- **THEN** SHALL 捕获数据库 UNIQUE 约束错误
+- **THEN** SHALL 转换为 HTTP 409 Conflict 错误返回
+- **THEN** SHALL 返回错误格式：
+  ```json
+  {
+    "error": "同一供应商下模型名称已存在",
+    "code": "duplicate_model"
+  }
+  ```
+
+#### Scenario: SQLite UNIQUE 约束错误检测
+
+- **WHEN** 捕获数据库错误
+- **THEN** SHALL 检查错误信息是否包含 "UNIQUE constraint failed"
+- **THEN** 如匹配，SHALL 识别为联合唯一约束冲突
 
 ### Requirement: 使用 repository 层访问数据
 
 Service SHALL 通过 ModelRepository 访问数据。
 
-#### Scenario: 调用 repository 方法
+#### Scenario: 联合查询
 
-- **WHEN** service 处理业务逻辑
+- **WHEN** service 需要按 provider 和 model_name 查询模型
-- **THEN** SHALL 调用对应的 ModelRepository 方法
+- **THEN** SHALL 调用 `FindByProviderAndModelName(providerID, modelName)` 方法
-- **THEN** SHALL 使用 domain.Model 类型
 
-#### Scenario: 数据验证
+#### Scenario: 查询所有启用模型
 
-- **WHEN** 创建或更新模型
+- **WHEN** proxy handler 需要聚合模型列表
-- **THEN** SHALL 在 service 层验证业务规则
+- **THEN** SHALL 调用 `ListEnabled()` 方法，返回所有 enabled 的模型（关联 enabled 的供应商）
-- **THEN** SHALL 在 repository 层执行数据库操作

openspec/specs/protocol-adapter-anthropic/spec.md

@@ -270,4 +270,73 @@ Decoder 几乎 1:1 映射，维护最小状态机：
 
 - **WHEN** interfaceType 为 EMBEDDINGS 或 RERANK
 - **THEN** supportsInterface SHALL 返回 false
-- **THEN** 引擎 SHALL 走透传或返回空响应
+- **THEN** 引擎 SHALL 走透传或返回空响应
+### Requirement: 模型详情路径识别
+
+Anthropic 适配器 SHALL 正确识别包含统一模型 ID 的模型详情路径。
+
+#### Scenario: 含斜杠的统一模型 ID 路径
+
+- **WHEN** 路径为 `/v1/models/anthropic/claude-3-opus`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
+
+#### Scenario: 含多段斜杠的统一模型 ID 路径
+
+- **WHEN** 路径为 `/v1/models/azure/deployments/gpt-4`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
+
+#### Scenario: 模型列表路径不受影响
+
+- **WHEN** 路径为 `/v1/models`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModels`
+
+### Requirement: 提取统一模型 ID
+
+Anthropic 适配器 SHALL 从路径中提取统一模型 ID。
+
+#### Scenario: 标准路径提取
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/anthropic/claude-3-opus")`
+- **THEN** SHALL 返回 `"anthropic/claude-3-opus"`
+
+#### Scenario: 复杂路径提取
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/azure/accounts/org/models/gpt-4")`
+- **THEN** SHALL 返回 `"azure/accounts/org/models/gpt-4"`
+
+#### Scenario: 非模型详情路径
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models")`
+- **THEN** SHALL 返回错误
+
+### Requirement: 从请求体提取 model
+
+Anthropic 适配器 SHALL 按 InterfaceType 从请求体中提取 model 值。
+
+#### Scenario: Chat 请求提取 model
+
+- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeChat)`，body 为 `{"model":"anthropic/claude-3-opus","messages":[...]}`
+- **THEN** SHALL 返回 `"anthropic/claude-3-opus"`
+
+#### Scenario: 无 model 字段
+
+- **WHEN** 调用 `ExtractModelName(body, ifaceType)`，body 中不含 model 字段
+- **THEN** SHALL 返回错误
+
+### Requirement: 最小化改写请求体 model 字段
+
+Anthropic 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写请求体中的 model 字段，其余字段保持原始 bytes。
+
+#### Scenario: 请求体 model 改写（统一 ID → 上游名）
+
+- **WHEN** 调用 `RewriteRequestModelName(body, "claude-3-opus-20240229", InterfaceTypeChat)`
+- **THEN** SHALL 将请求体中 model 字段替换为 `"claude-3-opus-20240229"`，其余字段原样保留
+
+### Requirement: 最小化改写响应体 model 字段
+
+Anthropic 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写响应体中的 model 字段，其余字段保持原始 bytes。请求体和响应体的 model 字段位置可能不同，各自独立实现。
+
+#### Scenario: 响应体 model 改写（上游名 → 统一 ID）
+
+- **WHEN** 调用 `RewriteResponseModelName(body, "anthropic/claude-3-opus", InterfaceTypeChat)`
+- **THEN** SHALL 将响应体中 model 字段替换为 `"anthropic/claude-3-opus"`，其余字段原样保留

openspec/specs/protocol-adapter-openai/spec.md

@@ -269,4 +269,91 @@ Encoder SHALL 维护状态：
 #### Scenario: /rerank 接口
 
 - **WHEN** 解码/编码 rerank 请求和响应
-- **THEN** SHALL 使用 CanonicalRerankRequest/Response 做字段映射
+- **THEN** SHALL 使用 CanonicalRerankRequest/Response 做字段映射
+### Requirement: 模型详情路径识别
+
+OpenAI 适配器 SHALL 正确识别包含统一模型 ID 的模型详情路径。
+
+#### Scenario: 含斜杠的统一模型 ID 路径
+
+- **WHEN** 路径为 `/v1/models/openai/gpt-4`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
+
+#### Scenario: 含多段斜杠的统一模型 ID 路径
+
+- **WHEN** 路径为 `/v1/models/azure/accounts/org-123/models/gpt-4`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModelInfo`
+
+#### Scenario: 模型列表路径不受影响
+
+- **WHEN** 路径为 `/v1/models`
+- **THEN** `DetectInterfaceType` SHALL 返回 `InterfaceTypeModels`
+
+### Requirement: 提取统一模型 ID
+
+OpenAI 适配器 SHALL 从路径中提取统一模型 ID。
+
+#### Scenario: 标准路径提取
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/openai/gpt-4")`
+- **THEN** SHALL 返回 `"openai/gpt-4"`
+
+#### Scenario: 复杂路径提取
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models/azure/accounts/org/models/gpt-4")`
+- **THEN** SHALL 返回 `"azure/accounts/org/models/gpt-4"`
+
+#### Scenario: 非模型详情路径
+
+- **WHEN** 调用 `ExtractUnifiedModelID("/v1/models")`
+- **THEN** SHALL 返回错误
+
+### Requirement: 从请求体提取 model
+
+OpenAI 适配器 SHALL 按 InterfaceType 从请求体中提取 model 值。
+
+#### Scenario: Chat 请求提取 model
+
+- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeChat)`，body 为 `{"model":"openai/gpt-4","messages":[...]}`
+- **THEN** SHALL 返回 `"openai/gpt-4"`
+
+#### Scenario: Embedding 请求提取 model
+
+- **WHEN** 调用 `ExtractModelName(body, InterfaceTypeEmbeddings)`，body 为 `{"model":"openai/text-embedding-3","input":"text"}`
+- **THEN** SHALL 返回 `"openai/text-embedding-3"`
+
+#### Scenario: 无 model 字段
+
+- **WHEN** 调用 `ExtractModelName(body, ifaceType)`，body 中不含 model 字段
+- **THEN** SHALL 返回错误
+
+### Requirement: 最小化改写请求体 model 字段
+
+OpenAI 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写请求体中的 model 字段，其余字段保持原始 bytes。
+
+#### Scenario: 请求体 model 改写（统一 ID → 上游名）
+
+- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeChat)`，body 为 `{"model":"openai/gpt-4","messages":[...],"some_param":"value"}`
+- **THEN** SHALL 返回 `{"model":"gpt-4","messages":[...],"some_param":"value"}`
+- **THEN** 除 model 外的字段 SHALL 保持原始 bytes 不变
+
+#### Scenario: 不同 InterfaceType 的请求改写
+
+- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeEmbeddings)`
+- **THEN** SHALL 按 Embedding 接口的请求体 model 字段位置进行改写
+- **WHEN** 调用 `RewriteRequestModelName(body, "gpt-4", InterfaceTypeRerank)`
+- **THEN** SHALL 按 Rerank 接口的请求体 model 字段位置进行改写
+
+### Requirement: 最小化改写响应体 model 字段
+
+OpenAI 适配器 SHALL 按 InterfaceType 用 `json.RawMessage` 最小化改写响应体中的 model 字段，其余字段保持原始 bytes。请求体和响应体的 model 字段位置可能不同，各自独立实现。
+
+#### Scenario: 响应体 model 改写（上游名 → 统一 ID）
+
+- **WHEN** 调用 `RewriteResponseModelName(body, "openai/gpt-4", InterfaceTypeChat)`，body 为上游 Chat 响应
+- **THEN** SHALL 将 model 字段替换为 `"openai/gpt-4"`，其余字段原样保留
+
+#### Scenario: 不同 InterfaceType 的响应改写
+
+- **WHEN** 调用 `RewriteResponseModelName(body, "openai/gpt-4", InterfaceTypeEmbeddings)`
+- **THEN** SHALL 按 Embedding 接口的响应体 model 字段位置进行改写

openspec/specs/provider-management/spec.md

@@ -29,7 +29,44 @@
 - **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
 - **THEN** 错误 SHALL 指示缺少哪些字段
 
-**变更说明：** handler 通过 ProviderService 调用，数据访问通过 ProviderRepository。API 接口保持不变。
+#### Scenario: 创建供应商时 ID 包含非法字符
+
+- **WHEN** 向 `/api/providers` 发送 POST 请求，id 包含非 `[a-zA-Z0-9_]` 字符
+- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
+- **THEN** 错误 SHALL 指示 id 仅允许字母、数字、下划线
+
+#### Scenario: 创建供应商时 ID 过长
+
+- **WHEN** 向 `/api/providers` 发送 POST 请求，id 长度超过 64
+- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
+
+### Requirement: 供应商 ID 不允许修改
+
+供应商 ID 是主键，用于构建统一模型 ID，不允许修改。
+
+#### Scenario: 尝试修改供应商 ID
+
+- **WHEN** 向 `/api/providers/:id` 发送 PUT 请求，请求体中包含 `id` 字段
+- **THEN** ProviderService.Update SHALL 在 service 层校验并返回错误
+- **THEN** 网关 SHALL 返回错误，状态码为 400 (Bad Request)
+- **THEN** 错误 SHALL 指示供应商 ID 不允许修改
+- **THEN** 错误格式 SHALL 为：
+  ```json
+  {
+    "error": "供应商 ID 不允许修改",
+    "code": "IMMUTABLE_FIELD"
+  }
+  ```
+
+**校验位置：** Service 层（`ProviderService.Update`）
+- Service 层校验保证所有调用方（handler、CLI、未来 API）统一遵守
+- Handler 层负责捕获 `ErrImmutableField` 并转换为 HTTP 400 响应
+
+**原因：**
+- 供应商 ID 是主键，用于构建统一模型 ID（provider_id/model_name）
+- 修改 ID 会导致所有统一模型 ID 失效
+- 客户端缓存的模型 ID 全部失效
+- 如需修改，应创建新供应商并迁移模型
 
 ### Requirement: 列出所有供应商
 

openspec/specs/request-validation/spec.md

@@ -92,6 +92,44 @@
 - **THEN** SHALL 验证至少提供一个可更新字段
 - **THEN** SHALL 验证字段值有效性
 
+### Requirement: 供应商 ID 校验
+
+创建供应商时，SHALL 对 `id` 字段进行字符集校验。
+
+#### Scenario: 合法字符集
+
+- **WHEN** 创建供应商，id 仅包含 `[a-zA-Z0-9_]` 字符
+- **THEN** SHALL 校验通过
+
+#### Scenario: 非法字符
+
+- **WHEN** 创建供应商，id 包含 `-`、`.`、`/`、空格、中文等非 `[a-zA-Z0-9_]` 字符
+- **THEN** SHALL 返回 400 错误
+
+#### Scenario: 长度限制
+
+- **WHEN** 创建供应商，id 长度超过 64
+- **THEN** SHALL 返回 400 错误
+
+### Requirement: 模型创建校验
+
+创建模型时，SHALL 对 `provider_id` + `model_name` 进行联合唯一性校验。
+
+#### Scenario: 正常创建
+
+- **WHEN** 创建模型，provider_id 存在且 provider_id + model_name 组合唯一
+- **THEN** SHALL 校验通过
+
+#### Scenario: 联合唯一冲突
+
+- **WHEN** 创建模型，provider_id + model_name 组合已存在
+- **THEN** SHALL 返回 409 错误
+
+#### Scenario: model_name 为空
+
+- **WHEN** 创建模型，未提供 model_name
+- **THEN** SHALL 返回 400 错误
+
 ### Requirement: 返回友好的验证错误
 
 系统 SHALL 返回友好的验证错误响应。

openspec/specs/unified-model-id/spec.md

@@ -1,4 +1,10 @@
-## ADDED Requirements
+# Unified Model ID
+
+## Purpose
+
+定义统一模型 ID 的格式、解析、格式化和校验规则，确保跨协议的模型标识一致性。
+
+## Requirements
 
 ### Requirement: 解析统一模型 ID
 

openspec/specs/unified-proxy-handler/spec.md

@@ -105,4 +105,125 @@ ProxyHandler SHALL 支持 GET 请求的扩展层接口代理。
 - **THEN** SHALL 调用 engine.convertHttpRequest（GET 请求 body 为空）
 - **THEN** SHALL 调用 providerClient.Send 发送请求
 - **THEN** SHALL 调用 engine.convertHttpResponse 转换响应格式
-- **THEN** SHALL 返回转换后的响应
+- **THEN** SHALL 返回转换后的响应
+### Requirement: 代理请求路由
+
+ProxyHandler SHALL 使用统一模型 ID 路由所有代理请求。
+
+#### Scenario: 提取统一模型 ID
+
+- **WHEN** 收到 Chat、Embeddings 或 Rerank 接口的 POST 请求（含请求体）
+- **THEN** SHALL 调用客户端协议 adapter 的 `ExtractModelName(body, ifaceType)` 提取 model 值
+- **THEN** SHALL 调用 `ParseUnifiedModelID` 解析得到 providerID 和 modelName
+- **THEN** SHALL 调用 `RoutingService.RouteByModelName(providerID, modelName)` 路由
+
+#### Scenario: GET 请求或无请求体
+
+- **WHEN** 收到 GET 请求或请求体为空或请求体中无法提取 model 字段
+- **THEN** SHALL 走 forwardPassthrough 透传到上游供应商（兼容未适配的客户端和无 body 请求）
+
+#### Scenario: 无效的统一模型 ID
+
+- **WHEN** 请求体中 `model` 字段不是有效的统一模型 ID 格式（不含 `/`）
+- **THEN** SHALL 走 forwardPassthrough 透传到上游供应商（兼容使用原始模型名的客户端）
+
+#### Scenario: 模型不存在
+
+- **WHEN** 解析统一模型 ID 后，数据库中找不到对应的 provider_id + model_name 组合
+- **THEN** SHALL 返回错误响应，状态码为 404
+
+#### Scenario: 模型已禁用
+
+- **WHEN** 解析统一模型 ID 后，对应的模型 enabled 为 false
+- **THEN** SHALL 返回错误响应，状态码为 404
+
+#### Scenario: 供应商已禁用
+
+- **WHEN** 解析统一模型 ID 后，对应的供应商 enabled 为 false
+- **THEN** SHALL 返回错误响应，状态码为 404
+
+### Requirement: 同协议 Smart Passthrough
+
+当客户端协议与供应商协议相同时，ProxyHandler SHALL 使用 Smart Passthrough 处理 Chat、Embedding、Rerank 请求。
+
+#### Scenario: 同协议非流式请求
+
+- **WHEN** 客户端协议 == 供应商协议，且为非流式请求
+- **THEN** SHALL 调用 adapter 的 `RewriteRequestModelName(body, modelName, ifaceType)` 将请求体中 model 从统一 ID 改写为上游模型名
+- **THEN** SHALL 构建 URL 和 Headers（同当前透传逻辑）
+- **THEN** SHALL 发送改写后的请求体到上游
+- **THEN** SHALL 调用 adapter 的 `RewriteResponseModelName(resp.Body, unifiedModelID, ifaceType)` 将响应中 model 从上游名改写为统一 ID
+- **THEN** SHALL NOT 对 body 做全量 decode → encode，保持未改写字段的原始 bytes
+
+#### Scenario: 同协议流式请求
+
+- **WHEN** 客户端协议 == 供应商协议，且为流式请求
+- **THEN** SHALL 对请求体做 `RewriteRequestModelName` 改写 model 字段
+- **THEN** SHALL 逐 SSE chunk 调用 `RewriteResponseModelName` 改写响应中 model 字段
+- **THEN** SHALL NOT 对 chunk 做全量 decode → encode
+
+#### Scenario: Smart Passthrough 保真性
+
+- **WHEN** 客户端发送含未知参数的请求（如 `{"model":"openai/gpt-4","some_new_param":"value"}`）
+- **THEN** 上游 SHALL 收到 `{"model":"gpt-4","some_new_param":"value"}`
+- **THEN** `some_new_param` SHALL 保持原始值不变，不丢失、不改变类型
+
+### Requirement: 跨协议完整转换
+
+当客户端协议与供应商协议不同时，ProxyHandler SHALL 使用全量转换路径。
+
+#### Scenario: 跨协议非流式请求
+
+- **WHEN** 客户端协议 != 供应商协议
+- **THEN** SHALL 走 `ConvertHttpRequest` 全量转换，encoder 中 provider.ModelName 覆盖 model
+- **THEN** SHALL 走 `ConvertHttpResponse` 全量转换，modelOverride 参数覆写 canonical.Model
+
+#### Scenario: 跨协议流式请求
+
+- **WHEN** 客户端协议 != 供应商协议，且为流式请求
+- **THEN** SHALL 走 `CreateStreamConverter` 全量转换，modelOverride 参数覆写流式 canonical 事件中的 Model
+
+### Requirement: 模型列表本地聚合
+
+ProxyHandler SHALL 从数据库聚合返回模型列表，不再透传上游。
+
+#### Scenario: GET /v1/models
+
+- **WHEN** 收到 `GET /{protocol}/v1/models` 请求
+- **THEN** SHALL 从数据库查询所有 enabled 的模型（关联 enabled 的供应商）
+- **THEN** SHALL 组装 `CanonicalModelList`，每个模型的 ID 字段为统一模型 ID（`provider_id/model_name`），Name 字段为 model_name，OwnedBy 字段为 provider_id
+- **THEN** SHALL 使用客户端协议的 adapter 编码响应
+- **THEN** SHALL NOT 请求上游供应商
+
+#### Scenario: 无可用模型
+
+- **WHEN** 数据库中没有 enabled 的模型
+- **THEN** SHALL 返回空列表
+
+### Requirement: 模型详情本地查询
+
+ProxyHandler SHALL 从数据库查询返回模型详情，不再透传上游。
+
+#### Scenario: GET /v1/models/{unified_id}
+
+- **WHEN** 收到 `GET /{protocol}/v1/models/{provider_id}/{model_name}` 请求
+- **THEN** SHALL 调用 adapter 的 `ExtractUnifiedModelID` 提取统一模型 ID
+- **THEN** SHALL 解析统一模型 ID 得到 providerID 和 modelName
+- **THEN** SHALL 从数据库查询对应的模型和供应商
+- **THEN** SHALL 组装 `CanonicalModelInfo`，ID 字段为统一模型 ID（`provider_id/model_name`），Name 字段为 model_name，OwnedBy 字段为 provider_id
+- **THEN** SHALL 使用客户端协议的 adapter 编码响应
+- **THEN** SHALL NOT 请求上游供应商
+
+#### Scenario: 模型详情不存在
+
+- **WHEN** 统一模型 ID 对应的模型不存在或已禁用
+- **THEN** SHALL 返回错误响应，状态码为 404
+
+### Requirement: 统计记录
+
+ProxyHandler SHALL 使用 providerID 和 modelName 记录使用统计。
+
+#### Scenario: 异步记录统计
+
+- **WHEN** 代理请求成功完成
+- **THEN** SHALL 异步调用 `StatsService.Record(providerID, modelName)`

openspec/specs/usage-statistics/spec.md

@@ -22,7 +22,20 @@
 - **THEN** 网关 SHALL 增加该供应商和模型的请求计数
 - **THEN** 网关 SHALL 在流结束后记录统计
 
-**变更说明：** 统计记录通过 StatsService 调用，数据访问通过 StatsRepository。API 接口保持不变。
+### Requirement: 使用统计记录统一模型标识
+
+系统 SHALL 使用 providerID 和 modelName（上游模型名）记录使用统计。
+
+#### Scenario: 代理请求统计记录
+
+- **WHEN** 代理请求成功完成
+- **THEN** SHALL 记录 provider_id 和 model_name 到 usage_stats 表（参数来自路由结果）
+- **THEN** SHALL 异步执行，不阻塞响应
+
+#### Scenario: 查询统计
+
+- **WHEN** 查询统计数据
+- **THEN** 支持按 provider_id 和 model_name 过滤
 
 ### Requirement: 按供应商查询统计