Merge origin/main into nightly

Resolve conflicts:
- .gitignore: keep nightly additions (.test, skills-lock.json)
- relay/helper/price.go: keep both billingexpr and model imports
- en.json / zh-CN.json: keep nightly's superset of i18n entries
- service/billing_session.go: add missing 3rd arg to DecreaseUserQuota
- en.json / zh-CN.json: deduplicate 129+320 duplicate i18n keys

relay/channel/aws/constants.go

@@ -18,6 +18,7 @@ var awsModelIDMap = map[string]string{
 	"claude-haiku-4-5-20251001":  "anthropic.claude-haiku-4-5-20251001-v1:0",
 	"claude-opus-4-5-20251101":   "anthropic.claude-opus-4-5-20251101-v1:0",
 	"claude-opus-4-6":            "anthropic.claude-opus-4-6-v1",
+	"claude-opus-4-7":            "anthropic.claude-opus-4-7",
 	// Nova models
 	"nova-micro-v1:0":   "amazon.nova-micro-v1:0",
 	"nova-lite-v1:0":    "amazon.nova-lite-v1:0",
@@ -91,6 +92,11 @@ var awsModelCanCrossRegionMap = map[string]map[string]bool{
 		"ap": true,
 		"eu": true,
 	},
+	"anthropic.claude-opus-4-7": {
+		"us": true,
+		"ap": true,
+		"eu": true,
+	},
 	"anthropic.claude-haiku-4-5-20251001-v1:0": {
 		"us": true,
 		"ap": true,

relay/channel/claude/constants.go

@@ -26,6 +26,13 @@ var ModelList = []string{
 	"claude-opus-4-6-medium",
 	"claude-opus-4-6-low",
 	"claude-sonnet-4-6",
+	"claude-opus-4-7",
+	"claude-opus-4-7-max",
+	"claude-opus-4-7-xhigh",
+	"claude-opus-4-7-high",
+	"claude-opus-4-7-medium",
+	"claude-opus-4-7-low",
+	"claude-opus-4-7-thinking",
 }
 
 var ChannelName = "claude"

relay/channel/claude/relay-claude.go

@@ -154,33 +154,52 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 	}
 
 	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(textRequest.Model); ok && effortLevel != "" &&
-		strings.HasPrefix(textRequest.Model, "claude-opus-4-6") {
+		(strings.HasPrefix(textRequest.Model, "claude-opus-4-6") || strings.HasPrefix(textRequest.Model, "claude-opus-4-7")) {
 		claudeRequest.Model = baseModel
 		claudeRequest.Thinking = &dto.Thinking{
 			Type: "adaptive",
 		}
 		claudeRequest.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
-		claudeRequest.TopP = common.GetPointer[float64](0)
-		claudeRequest.Temperature = common.GetPointer[float64](1.0)
+		if strings.HasPrefix(baseModel, "claude-opus-4-7") {
+			// Opus 4.7 rejects non-default temperature/top_p/top_k with 400
+			// and defaults display to "omitted"; restore the 4.6 visible summary.
+			claudeRequest.Thinking.Display = "summarized"
+			claudeRequest.Temperature = nil
+			claudeRequest.TopP = nil
+			claudeRequest.TopK = nil
+		} else {
+			claudeRequest.TopP = nil
+			claudeRequest.Temperature = common.GetPointer[float64](1.0)
+		}
 	} else if model_setting.GetClaudeSettings().ThinkingAdapterEnabled &&
 		strings.HasSuffix(textRequest.Model, "-thinking") {
 
-		// 因为BudgetTokens 必须大于1024
-		if claudeRequest.MaxTokens == nil || *claudeRequest.MaxTokens < 1280 {
-			claudeRequest.MaxTokens = common.GetPointer[uint](1280)
-		}
+		trimmedModel := strings.TrimSuffix(textRequest.Model, "-thinking")
+		if strings.HasPrefix(trimmedModel, "claude-opus-4-7") {
+			// Opus 4.7 rejects thinking.type="enabled"; use adaptive at high effort.
+			claudeRequest.Thinking = &dto.Thinking{Type: "adaptive", Display: "summarized"}
+			claudeRequest.OutputConfig = json.RawMessage(`{"effort":"high"}`)
+			claudeRequest.Temperature = nil
+			claudeRequest.TopP = nil
+			claudeRequest.TopK = nil
+		} else {
+			// 因为BudgetTokens 必须大于1024
+			if claudeRequest.MaxTokens == nil || *claudeRequest.MaxTokens < 1280 {
+				claudeRequest.MaxTokens = common.GetPointer[uint](1280)
+			}
 
-		// BudgetTokens 为 max_tokens 的 80%
-		claudeRequest.Thinking = &dto.Thinking{
-			Type:         "enabled",
-			BudgetTokens: common.GetPointer[int](int(float64(*claudeRequest.MaxTokens) * model_setting.GetClaudeSettings().ThinkingAdapterBudgetTokensPercentage)),
+			// BudgetTokens 为 max_tokens 的 80%
+			claudeRequest.Thinking = &dto.Thinking{
+				Type:         "enabled",
+				BudgetTokens: common.GetPointer[int](int(float64(*claudeRequest.MaxTokens) * model_setting.GetClaudeSettings().ThinkingAdapterBudgetTokensPercentage)),
+			}
+			// TODO: 临时处理
+			// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
+			claudeRequest.TopP = nil
+			claudeRequest.Temperature = common.GetPointer[float64](1.0)
 		}
-		// TODO: 临时处理
-		// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
-		claudeRequest.TopP = nil
-		claudeRequest.Temperature = common.GetPointer[float64](1.0)
 		if !model_setting.ShouldPreserveThinkingSuffix(textRequest.Model) {
-			claudeRequest.Model = strings.TrimSuffix(textRequest.Model, "-thinking")
+			claudeRequest.Model = trimmedModel
 		}
 	}
 
@@ -258,7 +277,7 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 				formatMessages = formatMessages[:len(formatMessages)-1]
 			}
 		}
-		if fmtMessage.Content == nil {
+		if fmtMessage.Content == nil || (fmtMessage.IsStringContent() && fmtMessage.StringContent() == "") {
 			fmtMessage.SetStringContent("...")
 		}
 		formatMessages = append(formatMessages, fmtMessage)
@@ -274,14 +293,16 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 		if message.Role == "system" {
 			// 根据Claude API规范，system字段使用数组格式更有通用性
 			if message.IsStringContent() {
-				systemMessages = append(systemMessages, dto.ClaudeMediaMessage{
-					Type: "text",
-					Text: common.GetPointer[string](message.StringContent()),
-				})
+				if text := message.StringContent(); text != "" {
+					systemMessages = append(systemMessages, dto.ClaudeMediaMessage{
+						Type: "text",
+						Text: common.GetPointer[string](text),
+					})
+				}
 			} else {
 				// 支持复合内容的system消息（虽然不常见，但需要考虑完整性）
 				for _, ctx := range message.ParseContent() {
-					if ctx.Type == "text" {
+					if ctx.Type == "text" && ctx.Text != "" {
 						systemMessages = append(systemMessages, dto.ClaudeMediaMessage{
 							Type: "text",
 							Text: common.GetPointer[string](ctx.Text),
@@ -339,16 +360,22 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 					}
 				}
 			} else if message.IsStringContent() && message.ToolCalls == nil {
-				claudeMessage.Content = message.StringContent()
+				text := message.StringContent()
+				if text == "" {
+					text = "..."
+				}
+				claudeMessage.Content = text
 			} else {
 				claudeMediaMessages := make([]dto.ClaudeMediaMessage, 0)
 				for _, mediaMessage := range message.ParseContent() {
 					switch mediaMessage.Type {
 					case "text":
-						claudeMediaMessages = append(claudeMediaMessages, dto.ClaudeMediaMessage{
-							Type: "text",
-							Text: common.GetPointer[string](mediaMessage.Text),
-						})
+						if mediaMessage.Text != "" {
+							claudeMediaMessages = append(claudeMediaMessages, dto.ClaudeMediaMessage{
+								Type: "text",
+								Text: common.GetPointer[string](mediaMessage.Text),
+							})
+						}
 					default:
 						source := mediaMessage.ToFileSource()
 						if source == nil {

relay/channel/openai/adaptor.go

@@ -136,8 +136,8 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			task = "chat/completions" + task
 		}
 
-		// 特殊处理 responses API
-		if info.RelayMode == relayconstant.RelayModeResponses {
+		// 特殊处理 responses API（包含 compact）
+		if info.RelayMode == relayconstant.RelayModeResponses || info.RelayMode == relayconstant.RelayModeResponsesCompact {
 			responsesApiVersion := "preview"
 
 			subUrl := "/openai/v1/responses"
@@ -150,6 +150,11 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 				responsesApiVersion = info.ChannelOtherSettings.AzureResponsesVersion
 			}
 
+			// compact 模式追加 /compact
+			if info.RelayMode == relayconstant.RelayModeResponsesCompact {
+				subUrl = subUrl + "/compact"
+			}
+
 			requestURL = fmt.Sprintf("%s?api-version=%s", subUrl, responsesApiVersion)
 			return relaycommon.GetFullRequestURL(info.ChannelBaseUrl, requestURL, info.ChannelType), nil
 		}

relay/channel/vertex/adaptor.go

@@ -44,6 +44,7 @@ var claudeModelMap = map[string]string{
 	"claude-haiku-4-5-20251001":  "claude-haiku-4-5@20251001",
 	"claude-opus-4-5-20251101":   "claude-opus-4-5@20251101",
 	"claude-opus-4-6":            "claude-opus-4-6",
+	"claude-opus-4-7":            "claude-opus-4-7",
 }
 
 const anthropicVersion = "vertex-2023-10-16"

relay/claude_handler.go

@@ -53,30 +53,49 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 	}
 
 	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(request.Model); ok && effortLevel != "" &&
-		strings.HasPrefix(request.Model, "claude-opus-4-6") {
+		(strings.HasPrefix(request.Model, "claude-opus-4-6") || strings.HasPrefix(request.Model, "claude-opus-4-7")) {
 		request.Model = baseModel
 		request.Thinking = &dto.Thinking{
 			Type: "adaptive",
 		}
 		request.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
-		request.Temperature = common.GetPointer[float64](1.0)
+		if strings.HasPrefix(request.Model, "claude-opus-4-7") {
+			// Opus 4.7 rejects non-default temperature/top_p/top_k with 400
+			// and defaults display to "omitted"; restore the 4.6 visible summary.
+			request.Thinking.Display = "summarized"
+			request.Temperature = nil
+			request.TopP = nil
+			request.TopK = nil
+		} else {
+			request.Temperature = common.GetPointer[float64](1.0)
+		}
 		info.UpstreamModelName = request.Model
 	} else if model_setting.GetClaudeSettings().ThinkingAdapterEnabled &&
 		strings.HasSuffix(request.Model, "-thinking") {
 		if request.Thinking == nil {
-			// 因为BudgetTokens 必须大于1024
-			if request.MaxTokens == nil || *request.MaxTokens < 1280 {
-				request.MaxTokens = common.GetPointer[uint](1280)
-			}
+			baseModel := strings.TrimSuffix(request.Model, "-thinking")
+			if strings.HasPrefix(baseModel, "claude-opus-4-7") {
+				// Opus 4.7 rejects thinking.type="enabled"; use adaptive at high effort.
+				request.Thinking = &dto.Thinking{Type: "adaptive", Display: "summarized"}
+				request.OutputConfig = json.RawMessage(`{"effort":"high"}`)
+				request.Temperature = nil
+				request.TopP = nil
+				request.TopK = nil
+			} else {
+				// 因为BudgetTokens 必须大于1024
+				if request.MaxTokens == nil || *request.MaxTokens < 1280 {
+					request.MaxTokens = common.GetPointer[uint](1280)
+				}
 
-			// BudgetTokens 为 max_tokens 的 80%
-			request.Thinking = &dto.Thinking{
-				Type:         "enabled",
-				BudgetTokens: common.GetPointer[int](int(float64(*request.MaxTokens) * model_setting.GetClaudeSettings().ThinkingAdapterBudgetTokensPercentage)),
+				// BudgetTokens 为 max_tokens 的 80%
+				request.Thinking = &dto.Thinking{
+					Type:         "enabled",
+					BudgetTokens: common.GetPointer[int](int(float64(*request.MaxTokens) * model_setting.GetClaudeSettings().ThinkingAdapterBudgetTokensPercentage)),
+				}
+				// TODO: 临时处理
+				// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
+				request.Temperature = common.GetPointer[float64](1.0)
 			}
-			// TODO: 临时处理
-			// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
-			request.Temperature = common.GetPointer[float64](1.0)
 		}
 		if !model_setting.ShouldPreserveThinkingSuffix(info.OriginModelName) {
 			request.Model = strings.TrimSuffix(request.Model, "-thinking")

relay/common/override.go

@@ -32,6 +32,7 @@ var paramOverrideKeyAuditPaths = map[string]struct{}{
 	"upstream_model": {},
 	"service_tier":   {},
 	"inference_geo":  {},
+	"speed":          {},
 }
 
 type paramOverrideAuditRecorder struct {

relay/common/override_test.go

@@ -2038,6 +2038,8 @@ func TestRemoveDisabledFieldsDefaultFiltering(t *testing.T) {
 	input := `{
 		"service_tier":"flex",
 		"inference_geo":"eu",
+		"speed":"fast",
+		"cache_control":{"type":"ephemeral"},
 		"safety_identifier":"user-123",
 		"store":true,
 		"stream_options":{"include_obfuscation":false}
@@ -2048,7 +2050,7 @@ func TestRemoveDisabledFieldsDefaultFiltering(t *testing.T) {
 	if err != nil {
 		t.Fatalf("RemoveDisabledFields returned error: %v", err)
 	}
-	assertJSONEqual(t, `{"store":true}`, string(out))
+	assertJSONEqual(t, `{"cache_control":{"type":"ephemeral"},"store":true}`, string(out))
 }
 
 func TestRemoveDisabledFieldsAllowInferenceGeo(t *testing.T) {
@@ -2067,6 +2069,22 @@ func TestRemoveDisabledFieldsAllowInferenceGeo(t *testing.T) {
 	assertJSONEqual(t, `{"inference_geo":"eu","store":true}`, string(out))
 }
 
+func TestRemoveDisabledFieldsAllowSpeed(t *testing.T) {
+	input := `{
+		"speed":"fast",
+		"store":true
+	}`
+	settings := dto.ChannelOtherSettings{
+		AllowSpeed: true,
+	}
+
+	out, err := RemoveDisabledFields([]byte(input), settings, false)
+	if err != nil {
+		t.Fatalf("RemoveDisabledFields returned error: %v", err)
+	}
+	assertJSONEqual(t, `{"speed":"fast","store":true}`, string(out))
+}
+
 func TestApplyParamOverrideWithRelayInfoRecordsOperationAuditInDebugMode(t *testing.T) {
 	originalDebugEnabled := common2.DebugEnabled
 	common2.DebugEnabled = true

relay/common/relay_info.go

@@ -444,6 +444,7 @@ func genBaseRelayInfo(c *gin.Context, request dto.Request) *RelayInfo {
 	if request != nil {
 		isStream = request.IsStream(c)
 	}
+	c.Set(string(constant.ContextKeyIsStream), isStream)
 
 	// firstResponseTime = time.Now() - 1 second
 
@@ -776,6 +777,7 @@ func FailTaskInfo(reason string) *TaskInfo {
 // RemoveDisabledFields 从请求 JSON 数据中移除渠道设置中禁用的字段
 // service_tier: 服务层级字段，可能导致额外计费（OpenAI、Claude、Responses API 支持）
 // inference_geo: Claude 数据驻留推理区域字段（仅 Claude 支持，默认过滤）
+// speed: Claude 推理速度模式字段（仅 Claude 支持，默认过滤）
 // store: 数据存储授权字段，涉及用户隐私（仅 OpenAI、Responses API 支持，默认允许透传，禁用后可能导致 Codex 无法使用）
 // safety_identifier: 安全标识符，用于向 OpenAI 报告违规用户（仅 OpenAI 支持，涉及用户隐私）
 // stream_options.include_obfuscation: 响应流混淆控制字段（仅 OpenAI Responses API 支持）
@@ -804,6 +806,13 @@ func RemoveDisabledFields(jsonData []byte, channelOtherSettings dto.ChannelOther
 		}
 	}
 
+	// 默认移除 speed，除非明确允许（避免意外切换 Claude 推理速度模式）
+	if !channelOtherSettings.AllowSpeed {
+		if _, exists := data["speed"]; exists {
+			delete(data, "speed")
+		}
+	}
+
 	// 默认允许 store 透传，除非明确禁用（禁用可能影响 Codex 使用）
 	if channelOtherSettings.DisableStore {
 		if _, exists := data["store"]; exists {

relay/helper/price.go

@@ -5,6 +5,7 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/logger"
+	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/pkg/billingexpr"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/setting/billing_setting"
@@ -15,6 +16,21 @@ import (
 	"github.com/gin-gonic/gin"
 )
 
+func modelPriceNotConfiguredError(modelName string, userId int) error {
+	if model.IsAdmin(userId) {
+		return fmt.Errorf(
+			"模型 %s 的价格未配置。请前往「系统设置 → 运营设置」开启自用模式，或在「系统设置 → 分组与模型定价设置」中为该模型配置价格；"+
+				"Model %s price not configured. Go to System Settings → Operation Settings to enable self-use mode, or configure the model price in System Settings → Group & Model Pricing.",
+			modelName, modelName,
+		)
+	}
+	return fmt.Errorf(
+		"模型 %s 的价格尚未由管理员配置，暂时无法使用，请联系站点管理员开启该模型；"+
+			"Model %s has not been priced by the administrator yet. Please contact the site administrator to enable this model.",
+		modelName, modelName,
+	)
+}
+
 // https://docs.claude.com/en/docs/build-with-claude/prompt-caching#1-hour-cache-duration
 const claudeCacheCreation1hMultiplier = 6 / 3.75
 
@@ -82,7 +98,7 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 				acceptUnsetRatio = true
 			}
 			if !acceptUnsetRatio {
-				return types.PriceData{}, fmt.Errorf("模型 %s 倍率或价格未配置，请联系管理员设置或开始自用模式；Model %s ratio or price not set, please set or start self-use mode", matchName, matchName)
+				return types.PriceData{}, modelPriceNotConfiguredError(matchName, info.UserId)
 			}
 		}
 		completionRatio = ratio_setting.GetCompletionRatio(info.OriginModelName)
@@ -168,7 +184,7 @@ func ModelPriceHelperPerCall(c *gin.Context, info *relaycommon.RelayInfo) (types
 				acceptUnsetRatio = true
 			}
 			if !ratioSuccess && !acceptUnsetRatio {
-				return types.PriceData{}, fmt.Errorf("模型 %s 倍率或价格未配置，请联系管理员设置或开始自用模式；Model %s ratio or price not set, please set or start self-use mode", matchName, matchName)
+				return types.PriceData{}, modelPriceNotConfiguredError(matchName, info.UserId)
 			}
 		}
 	}

relay/responses_handler.go

@@ -143,7 +143,7 @@ func ResponsesHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 		if err != nil {
 			info.OriginModelName = originModelName
 			info.PriceData = originPriceData
-			return types.NewError(err, types.ErrorCodeModelPriceError, types.ErrOptionWithSkipRetry())
+			return types.NewError(err, types.ErrorCodeModelPriceError, types.ErrOptionWithSkipRetry(), types.ErrOptionWithStatusCode(http.StatusBadRequest))
 		}
 		service.PostTextConsumeQuota(c, info, usageDto, nil)