Merge pull request #2863 from prnake/feat/claude-opus-4-6

Calcium-Ion · web-flow · commit f3d6e99b2868 · 2026-02-06T16:18:00.000+08:00
feat: add claude-opus-4-6
diff --git a/relay/channel/aws/constants.go b/relay/channel/aws/constants.go
@@ -3,7 +3,7 @@ package aws
 import "strings"
 
 var awsModelIDMap = map[string]string{
-	"claude-3-sonnet-20240229": "anthropic.claude-3-sonnet-20240229-v1:0",
+	"claude-3-sonnet-20240229":   "anthropic.claude-3-sonnet-20240229-v1:0",
 	"claude-3-opus-20240229":     "anthropic.claude-3-opus-20240229-v1:0",
 	"claude-3-haiku-20240307":    "anthropic.claude-3-haiku-20240307-v1:0",
 	"claude-3-5-sonnet-20240620": "anthropic.claude-3-5-sonnet-20240620-v1:0",
@@ -16,6 +16,7 @@ var awsModelIDMap = map[string]string{
 	"claude-sonnet-4-5-20250929": "anthropic.claude-sonnet-4-5-20250929-v1:0",
 	"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",
 	// Nova models
 	"nova-micro-v1:0":   "amazon.nova-micro-v1:0",
 	"nova-lite-v1:0":    "amazon.nova-lite-v1:0",
@@ -79,6 +80,11 @@ var awsModelCanCrossRegionMap = map[string]map[string]bool{
 		"ap": true,
 		"eu": true,
 	},
+	"anthropic.claude-opus-4-6-v1": {
+		"us": true,
+		"ap": true,
+		"eu": true,
+	},
 	"anthropic.claude-haiku-4-5-20251001-v1:0": {
 		"us": true,
 		"ap": true,
diff --git a/relay/channel/aws/dto.go b/relay/channel/aws/dto.go
@@ -26,6 +26,7 @@ type AwsClaudeRequest struct {
 	Tools            any                 `json:"tools,omitempty"`
 	ToolChoice       any                 `json:"tool_choice,omitempty"`
 	Thinking         *dto.Thinking       `json:"thinking,omitempty"`
+	OutputConfig     json.RawMessage     `json:"output_config,omitempty"`
 }
 
 func formatRequest(requestBody io.Reader, requestHeader http.Header) (*AwsClaudeRequest, error) {
diff --git a/relay/channel/claude/constants.go b/relay/channel/claude/constants.go
@@ -20,6 +20,11 @@ var ModelList = []string{
 	"claude-sonnet-4-5-20250929-thinking",
 	"claude-opus-4-5-20251101",
 	"claude-opus-4-5-20251101-thinking",
+	"claude-opus-4-6",
+	"claude-opus-4-6-max",
+	"claude-opus-4-6-high",
+	"claude-opus-4-6-medium",
+	"claude-opus-4-6-low",
 }
 
 var ChannelName = "claude"
diff --git a/relay/channel/claude/relay-claude.go b/relay/channel/claude/relay-claude.go
@@ -17,6 +17,7 @@ import (
 	"github.com/QuantumNous/new-api/relay/reasonmap"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/model_setting"
+	"github.com/QuantumNous/new-api/setting/reasoning"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -141,7 +142,16 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 		claudeRequest.MaxTokens = uint(model_setting.GetClaudeSettings().GetDefaultMaxTokens(textRequest.Model))
 	}
 
-	if model_setting.GetClaudeSettings().ThinkingAdapterEnabled &&
+	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(textRequest.Model); ok && effortLevel != "" &&
+		strings.HasPrefix(textRequest.Model, "claude-opus-4-6") {
+		claudeRequest.Model = baseModel
+		claudeRequest.Thinking = &dto.Thinking{
+			Type: "adaptive",
+		}
+		claudeRequest.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
+		claudeRequest.TopP = 0
+		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
diff --git a/relay/channel/vertex/adaptor.go b/relay/channel/vertex/adaptor.go
@@ -42,6 +42,7 @@ var claudeModelMap = map[string]string{
 	"claude-sonnet-4-5-20250929": "claude-sonnet-4-5@20250929",
 	"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",
 }
 
 const anthropicVersion = "vertex-2023-10-16"
diff --git a/relay/channel/vertex/dto.go b/relay/channel/vertex/dto.go
@@ -1,6 +1,8 @@
 package vertex
 
 import (
+	"encoding/json"
+
 	"github.com/QuantumNous/new-api/dto"
 )
 
@@ -17,6 +19,7 @@ type VertexAIClaudeRequest struct {
 	Tools            any                 `json:"tools,omitempty"`
 	ToolChoice       any                 `json:"tool_choice,omitempty"`
 	Thinking         *dto.Thinking       `json:"thinking,omitempty"`
+	OutputConfig     json.RawMessage     `json:"output_config,omitempty"`
 }
 
 func copyRequest(req *dto.ClaudeRequest, version string) *VertexAIClaudeRequest {
@@ -33,5 +36,6 @@ func copyRequest(req *dto.ClaudeRequest, version string) *VertexAIClaudeRequest
 		Tools:            req.Tools,
 		ToolChoice:       req.ToolChoice,
 		Thinking:         req.Thinking,
+		OutputConfig:     req.OutputConfig,
 	}
 }
diff --git a/relay/claude_handler.go b/relay/claude_handler.go
@@ -2,6 +2,7 @@ package relay
 
 import (
 	"bytes"
+	"encoding/json"
 	"fmt"
 	"io"
 	"net/http"
@@ -14,6 +15,7 @@ import (
 	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/model_setting"
+	"github.com/QuantumNous/new-api/setting/reasoning"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
@@ -49,7 +51,17 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		request.MaxTokens = uint(model_setting.GetClaudeSettings().GetDefaultMaxTokens(request.Model))
 	}
 
-	if model_setting.GetClaudeSettings().ThinkingAdapterEnabled &&
+	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(request.Model); ok && effortLevel != "" &&
+		strings.HasPrefix(request.Model, "claude-opus-4-6") {
+		request.Model = baseModel
+		request.Thinking = &dto.Thinking{
+			Type: "adaptive",
+		}
+		request.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
+		request.TopP = 0
+		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
diff --git a/setting/ratio_setting/cache_ratio.go b/setting/ratio_setting/cache_ratio.go
@@ -60,6 +60,12 @@ var defaultCacheRatio = map[string]float64{
 	"claude-sonnet-4-5-20250929-thinking": 0.1,
 	"claude-opus-4-5-20251101":            0.1,
 	"claude-opus-4-5-20251101-thinking":   0.1,
+	"claude-opus-4-6":                     0.1,
+	"claude-opus-4-6-thinking":            0.1,
+	"claude-opus-4-6-max":                 0.1,
+	"claude-opus-4-6-high":                0.1,
+	"claude-opus-4-6-medium":              0.1,
+	"claude-opus-4-6-low":                 0.1,
 }
 
 var defaultCreateCacheRatio = map[string]float64{
@@ -82,6 +88,12 @@ var defaultCreateCacheRatio = map[string]float64{
 	"claude-sonnet-4-5-20250929-thinking": 1.25,
 	"claude-opus-4-5-20251101":            1.25,
 	"claude-opus-4-5-20251101-thinking":   1.25,
+	"claude-opus-4-6":                     1.25,
+	"claude-opus-4-6-thinking":            1.25,
+	"claude-opus-4-6-max":                 1.25,
+	"claude-opus-4-6-high":                1.25,
+	"claude-opus-4-6-medium":              1.25,
+	"claude-opus-4-6-low":                 1.25,
 }
 
 //var defaultCreateCacheRatio = map[string]float64{}
diff --git a/setting/ratio_setting/model_ratio.go b/setting/ratio_setting/model_ratio.go
@@ -142,6 +142,11 @@ var defaultModelRatio = map[string]float64{
 	"claude-sonnet-4-20250514":                  1.5,
 	"claude-sonnet-4-5-20250929":                1.5,
 	"claude-opus-4-5-20251101":                  2.5,
+	"claude-opus-4-6":                           2.5,
+	"claude-opus-4-6-max":                       2.5,
+	"claude-opus-4-6-high":                      2.5,
+	"claude-opus-4-6-medium":                    2.5,
+	"claude-opus-4-6-low":                       2.5,
 	"claude-3-opus-20240229":                    7.5, // $15 / 1M tokens
 	"claude-opus-4-20250514":                    7.5,
 	"claude-opus-4-1-20250805":                  7.5,
diff --git a/setting/reasoning/suffix.go b/setting/reasoning/suffix.go
@@ -6,7 +6,7 @@ import (
 	"github.com/samber/lo"
 )
 
-var EffortSuffixes = []string{"-high", "-medium", "-low", "-minimal"}
+var EffortSuffixes = []string{"-max", "-high", "-medium", "-low", "-minimal"}
 
 // TrimEffortSuffix -> modelName level(low) exists
 func TrimEffortSuffix(modelName string) (string, string, bool) {

Original file line number	Diff line number	Diff line change
`@@ -26,6 +26,7 @@ type AwsClaudeRequest struct {`
`26`	`26`	Tools any `json:"tools,omitempty"`
`27`	`27`	ToolChoice any `json:"tool_choice,omitempty"`
`28`	`28`	Thinking *dto.Thinking `json:"thinking,omitempty"`
	`29`	+ OutputConfig json.RawMessage `json:"output_config,omitempty"`
`29`	`30`	`}`
`30`	`31`
`31`	`32`	`func formatRequest(requestBody io.Reader, requestHeader http.Header) (*AwsClaudeRequest, error) {`
Original file line number	Diff line number	Diff line change
`@@ -42,6 +42,7 @@ var claudeModelMap = map[string]string{`
`42`	`42`	`"claude-sonnet-4-5-20250929": "claude-sonnet-4-5@20250929",`
`43`	`43`	`"claude-haiku-4-5-20251001": "claude-haiku-4-5@20251001",`
`44`	`44`	`"claude-opus-4-5-20251101": "claude-opus-4-5@20251101",`
	`45`	`+ "claude-opus-4-6": "claude-opus-4-6",`
`45`	`46`	`}`
`46`	`47`
`47`	`48`	`const anthropicVersion = "vertex-2023-10-16"`
Original file line number	Diff line number	Diff line change
`@@ -1,6 +1,8 @@`
`1`	`1`	`package vertex`
`2`	`2`
`3`	`3`	`import (`
	`4`	`+ "encoding/json"`
	`5`	`+`
`4`	`6`	`"github.com/QuantumNous/new-api/dto"`
`5`	`7`	`)`
`6`	`8`
`@@ -17,6 +19,7 @@ type VertexAIClaudeRequest struct {`
`17`	`19`	Tools any `json:"tools,omitempty"`
`18`	`20`	ToolChoice any `json:"tool_choice,omitempty"`
`19`	`21`	Thinking *dto.Thinking `json:"thinking,omitempty"`
	`22`	+ OutputConfig json.RawMessage `json:"output_config,omitempty"`
`20`	`23`	`}`
`21`	`24`
`22`	`25`	`func copyRequest(req dto.ClaudeRequest, version string) VertexAIClaudeRequest {`
`@@ -33,5 +36,6 @@ func copyRequest(req dto.ClaudeRequest, version string) VertexAIClaudeRequest`
`33`	`36`	`Tools: req.Tools,`
`34`	`37`	`ToolChoice: req.ToolChoice,`
`35`	`38`	`Thinking: req.Thinking,`
	`39`	`+ OutputConfig: req.OutputConfig,`
`36`	`40`	`}`
`37`	`41`	`}`
Original file line number	Diff line number	Diff line change
`@@ -6,7 +6,7 @@ import (`
`6`	`6`	`"github.com/samber/lo"`
`7`	`7`	`)`
`8`	`8`
`9`		`-var EffortSuffixes = []string{"-high", "-medium", "-low", "-minimal"}`
	`9`	`+var EffortSuffixes = []string{"-max", "-high", "-medium", "-low", "-minimal"}`
`10`	`10`
`11`	`11`	`// TrimEffortSuffix -> modelName level(low) exists`
`12`	`12`	`func TrimEffortSuffix(modelName string) (string, string, bool) {`