mirror of
https://github.com/router-for-me/CLIProxyAPI.git
synced 2026-02-04 05:20:52 +08:00
Compare commits
17 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
9e5b1d24e8 | ||
|
|
a7dae6ad52 | ||
|
|
e93e05ae25 | ||
|
c8c27325dc | ||
|
|
c3b6f3918c | ||
|
|
bbb55a8ab4 | ||
|
|
7583193c2a | ||
|
|
7cc3bd4ba0 | ||
|
|
88a0f095e8 | ||
|
|
c65f64dce0 | ||
|
|
d18cd217e1 | ||
|
|
ba4a1ab433 | ||
|
decddb521e | ||
|
|
70897247b2 | ||
|
|
9c341f5aa5 | ||
|
f7bfa8a05c | ||
|
|
c8620d1633 |
2
go.mod
2
go.mod
@@ -21,7 +21,6 @@ require (
|
||||
golang.org/x/crypto v0.45.0
|
||||
golang.org/x/net v0.47.0
|
||||
golang.org/x/oauth2 v0.30.0
|
||||
golang.org/x/text v0.31.0
|
||||
gopkg.in/natefinch/lumberjack.v2 v2.2.1
|
||||
gopkg.in/yaml.v3 v3.0.1
|
||||
)
|
||||
@@ -71,6 +70,7 @@ require (
|
||||
golang.org/x/arch v0.8.0 // indirect
|
||||
golang.org/x/sync v0.18.0 // indirect
|
||||
golang.org/x/sys v0.38.0 // indirect
|
||||
golang.org/x/text v0.31.0 // indirect
|
||||
google.golang.org/protobuf v1.34.1 // indirect
|
||||
gopkg.in/ini.v1 v1.67.0 // indirect
|
||||
)
|
||||
|
||||
33
internal/api/handlers/management/model_definitions.go
Normal file
33
internal/api/handlers/management/model_definitions.go
Normal file
@@ -0,0 +1,33 @@
|
||||
package management
|
||||
|
||||
import (
|
||||
"net/http"
|
||||
"strings"
|
||||
|
||||
"github.com/gin-gonic/gin"
|
||||
"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
|
||||
)
|
||||
|
||||
// GetStaticModelDefinitions returns static model metadata for a given channel.
|
||||
// Channel is provided via path param (:channel) or query param (?channel=...).
|
||||
func (h *Handler) GetStaticModelDefinitions(c *gin.Context) {
|
||||
channel := strings.TrimSpace(c.Param("channel"))
|
||||
if channel == "" {
|
||||
channel = strings.TrimSpace(c.Query("channel"))
|
||||
}
|
||||
if channel == "" {
|
||||
c.JSON(http.StatusBadRequest, gin.H{"error": "channel is required"})
|
||||
return
|
||||
}
|
||||
|
||||
models := registry.GetStaticModelDefinitionsByChannel(channel)
|
||||
if models == nil {
|
||||
c.JSON(http.StatusBadRequest, gin.H{"error": "unknown channel", "channel": channel})
|
||||
return
|
||||
}
|
||||
|
||||
c.JSON(http.StatusOK, gin.H{
|
||||
"channel": strings.ToLower(strings.TrimSpace(channel)),
|
||||
"models": models,
|
||||
})
|
||||
}
|
||||
@@ -607,6 +607,7 @@ func (s *Server) registerManagementRoutes() {
|
||||
|
||||
mgmt.GET("/auth-files", s.mgmt.ListAuthFiles)
|
||||
mgmt.GET("/auth-files/models", s.mgmt.GetAuthFileModels)
|
||||
mgmt.GET("/model-definitions/:channel", s.mgmt.GetStaticModelDefinitions)
|
||||
mgmt.GET("/auth-files/download", s.mgmt.DownloadAuthFile)
|
||||
mgmt.POST("/auth-files", s.mgmt.UploadAuthFile)
|
||||
mgmt.DELETE("/auth-files", s.mgmt.DeleteAuthFile)
|
||||
|
||||
@@ -1,848 +1,69 @@
|
||||
// Package registry provides model definitions for various AI service providers.
|
||||
// This file contains static model definitions that can be used by clients
|
||||
// when registering their supported models.
|
||||
// Package registry provides model definitions and lookup helpers for various AI providers.
|
||||
// Static model metadata is stored in model_definitions_static_data.go.
|
||||
package registry
|
||||
|
||||
// GetClaudeModels returns the standard Claude model definitions
|
||||
func GetClaudeModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
import (
|
||||
"sort"
|
||||
"strings"
|
||||
)
|
||||
|
||||
{
|
||||
ID: "claude-haiku-4-5-20251001",
|
||||
Object: "model",
|
||||
Created: 1759276800, // 2025-10-01
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Haiku",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
// Thinking: not supported for Haiku models
|
||||
},
|
||||
{
|
||||
ID: "claude-sonnet-4-5-20250929",
|
||||
Object: "model",
|
||||
Created: 1759104000, // 2025-09-29
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Sonnet",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-5-20251101",
|
||||
Object: "model",
|
||||
Created: 1761955200, // 2025-11-01
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Opus",
|
||||
Description: "Premium model combining maximum intelligence with practical performance",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-1-20250805",
|
||||
Object: "model",
|
||||
Created: 1722945600, // 2025-08-05
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.1 Opus",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 32000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-20250514",
|
||||
Object: "model",
|
||||
Created: 1715644800, // 2025-05-14
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4 Opus",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 32000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-sonnet-4-20250514",
|
||||
Object: "model",
|
||||
Created: 1715644800, // 2025-05-14
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4 Sonnet",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-3-7-sonnet-20250219",
|
||||
Object: "model",
|
||||
Created: 1708300800, // 2025-02-19
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 3.7 Sonnet",
|
||||
ContextLength: 128000,
|
||||
MaxCompletionTokens: 8192,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-3-5-haiku-20241022",
|
||||
Object: "model",
|
||||
Created: 1729555200, // 2024-10-22
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 3.5 Haiku",
|
||||
ContextLength: 128000,
|
||||
MaxCompletionTokens: 8192,
|
||||
// Thinking: not supported for Haiku models
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetGeminiModels returns the standard Gemini model definitions
|
||||
func GetGeminiModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Gemini 3 Flash Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-image-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-image-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Image Preview",
|
||||
Description: "Gemini 3 Pro Image Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
func GetGeminiVertexModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-image-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-image-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Image Preview",
|
||||
Description: "Gemini 3 Pro Image Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
// Imagen image generation models - use :predict action
|
||||
{
|
||||
ID: "imagen-4.0-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Generate",
|
||||
Description: "Imagen 4.0 image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-4.0-ultra-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-ultra-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Ultra Generate",
|
||||
Description: "Imagen 4.0 Ultra high-quality image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-3.0-generate-002",
|
||||
Object: "model",
|
||||
Created: 1740000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-3.0-generate-002",
|
||||
Version: "3.0",
|
||||
DisplayName: "Imagen 3.0 Generate",
|
||||
Description: "Imagen 3.0 image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-3.0-fast-generate-001",
|
||||
Object: "model",
|
||||
Created: 1740000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-3.0-fast-generate-001",
|
||||
Version: "3.0",
|
||||
DisplayName: "Imagen 3.0 Fast Generate",
|
||||
Description: "Imagen 3.0 fast image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-4.0-fast-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-fast-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Fast Generate",
|
||||
Description: "Imagen 4.0 fast image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetGeminiCLIModels returns the standard Gemini model definitions
|
||||
func GetGeminiCLIModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Our most intelligent model with SOTA reasoning and multimodal understanding, and powerful agentic and vibe coding capabilities",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetAIStudioModels returns the Gemini model definitions for AI Studio integrations
|
||||
func GetAIStudioModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-pro-latest",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-pro-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Pro Latest",
|
||||
Description: "Latest release of Gemini Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-flash-latest",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-flash-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Flash Latest",
|
||||
Description: "Latest release of Gemini Flash",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-flash-lite-latest",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-flash-lite-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Flash-Lite Latest",
|
||||
Description: "Latest release of Gemini Flash-Lite",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-image-preview",
|
||||
Object: "model",
|
||||
Created: 1756166400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-image-preview",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Image Preview",
|
||||
Description: "State-of-the-art image generation and editing model.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 8192,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
// image models don't support thinkingConfig; leave Thinking nil
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-image",
|
||||
Object: "model",
|
||||
Created: 1759363200,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-image",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Image",
|
||||
Description: "State-of-the-art image generation and editing model.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 8192,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
// image models don't support thinkingConfig; leave Thinking nil
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetOpenAIModels returns the standard OpenAI model definitions
|
||||
func GetOpenAIModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gpt-5",
|
||||
Object: "model",
|
||||
Created: 1754524800,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-08-07",
|
||||
DisplayName: "GPT 5",
|
||||
Description: "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5-codex",
|
||||
Object: "model",
|
||||
Created: 1757894400,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-09-15",
|
||||
DisplayName: "GPT 5 Codex",
|
||||
Description: "Stable version of GPT 5 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5-codex-mini",
|
||||
Object: "model",
|
||||
Created: 1762473600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-11-07",
|
||||
DisplayName: "GPT 5 Codex Mini",
|
||||
Description: "Stable version of GPT 5 Codex Mini: cheaper, faster, but less capable version of GPT 5 Codex.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5",
|
||||
Description: "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"none", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5.1 Codex",
|
||||
Description: "Stable version of GPT 5.1 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex-mini",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5.1 Codex Mini",
|
||||
Description: "Stable version of GPT 5.1 Codex Mini: cheaper, faster, but less capable version of GPT 5.1 Codex.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex-max",
|
||||
Object: "model",
|
||||
Created: 1763424000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-max",
|
||||
DisplayName: "GPT 5.1 Codex Max",
|
||||
Description: "Stable version of GPT 5.1 Codex Max",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.2",
|
||||
Object: "model",
|
||||
Created: 1765440000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.2",
|
||||
DisplayName: "GPT 5.2",
|
||||
Description: "Stable version of GPT 5.2",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"none", "low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.2-codex",
|
||||
Object: "model",
|
||||
Created: 1765440000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.2",
|
||||
DisplayName: "GPT 5.2 Codex",
|
||||
Description: "Stable version of GPT 5.2 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetQwenModels returns the standard Qwen model definitions
|
||||
func GetQwenModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "qwen3-coder-plus",
|
||||
Object: "model",
|
||||
Created: 1753228800,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Coder Plus",
|
||||
Description: "Advanced code generation and understanding model",
|
||||
ContextLength: 32768,
|
||||
MaxCompletionTokens: 8192,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
{
|
||||
ID: "qwen3-coder-flash",
|
||||
Object: "model",
|
||||
Created: 1753228800,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Coder Flash",
|
||||
Description: "Fast code generation model",
|
||||
ContextLength: 8192,
|
||||
MaxCompletionTokens: 2048,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
{
|
||||
ID: "vision-model",
|
||||
Object: "model",
|
||||
Created: 1758672000,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Vision Model",
|
||||
Description: "Vision model model",
|
||||
ContextLength: 32768,
|
||||
MaxCompletionTokens: 2048,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// iFlowThinkingSupport is a shared ThinkingSupport configuration for iFlow models
|
||||
// that support thinking mode via chat_template_kwargs.enable_thinking (boolean toggle).
|
||||
// Uses level-based configuration so standard normalization flows apply before conversion.
|
||||
var iFlowThinkingSupport = &ThinkingSupport{
|
||||
Levels: []string{"none", "auto", "minimal", "low", "medium", "high", "xhigh"},
|
||||
}
|
||||
|
||||
// GetIFlowModels returns supported models for iFlow OAuth accounts.
|
||||
func GetIFlowModels() []*ModelInfo {
|
||||
entries := []struct {
|
||||
ID string
|
||||
DisplayName string
|
||||
Description string
|
||||
Created int64
|
||||
Thinking *ThinkingSupport
|
||||
}{
|
||||
{ID: "tstars2.0", DisplayName: "TStars-2.0", Description: "iFlow TStars-2.0 multimodal assistant", Created: 1746489600},
|
||||
{ID: "qwen3-coder-plus", DisplayName: "Qwen3-Coder-Plus", Description: "Qwen3 Coder Plus code generation", Created: 1753228800},
|
||||
{ID: "qwen3-max", DisplayName: "Qwen3-Max", Description: "Qwen3 flagship model", Created: 1758672000},
|
||||
{ID: "qwen3-vl-plus", DisplayName: "Qwen3-VL-Plus", Description: "Qwen3 multimodal vision-language", Created: 1758672000},
|
||||
{ID: "qwen3-max-preview", DisplayName: "Qwen3-Max-Preview", Description: "Qwen3 Max preview build", Created: 1757030400},
|
||||
{ID: "kimi-k2-0905", DisplayName: "Kimi-K2-Instruct-0905", Description: "Moonshot Kimi K2 instruct 0905", Created: 1757030400},
|
||||
{ID: "glm-4.6", DisplayName: "GLM-4.6", Description: "Zhipu GLM 4.6 general model", Created: 1759190400, Thinking: iFlowThinkingSupport},
|
||||
{ID: "glm-4.7", DisplayName: "GLM-4.7", Description: "Zhipu GLM 4.7 general model", Created: 1766448000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "kimi-k2", DisplayName: "Kimi-K2", Description: "Moonshot Kimi K2 general model", Created: 1752192000},
|
||||
{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 thinking model", Created: 1762387200},
|
||||
{ID: "deepseek-v3.2-chat", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Chat", Created: 1764576000},
|
||||
{ID: "deepseek-v3.2-reasoner", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Reasoner", Created: 1764576000},
|
||||
{ID: "deepseek-v3.2", DisplayName: "DeepSeek-V3.2-Exp", Description: "DeepSeek V3.2 experimental", Created: 1759104000},
|
||||
{ID: "deepseek-v3.1", DisplayName: "DeepSeek-V3.1-Terminus", Description: "DeepSeek V3.1 Terminus", Created: 1756339200},
|
||||
{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200},
|
||||
{ID: "deepseek-v3", DisplayName: "DeepSeek-V3-671B", Description: "DeepSeek V3 671B", Created: 1734307200},
|
||||
{ID: "qwen3-32b", DisplayName: "Qwen3-32B", Description: "Qwen3 32B", Created: 1747094400},
|
||||
{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600},
|
||||
{ID: "qwen3-235b-a22b-instruct", DisplayName: "Qwen3-235B-A22B-Instruct", Description: "Qwen3 235B A22B Instruct", Created: 1753401600},
|
||||
{ID: "qwen3-235b", DisplayName: "Qwen3-235B-A22B", Description: "Qwen3 235B A22B", Created: 1753401600},
|
||||
{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "minimax-m2.1", DisplayName: "MiniMax-M2.1", Description: "MiniMax M2.1", Created: 1766448000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "iflow-rome-30ba3b", DisplayName: "iFlow-ROME", Description: "iFlow Rome 30BA3B model", Created: 1736899200},
|
||||
}
|
||||
models := make([]*ModelInfo, 0, len(entries))
|
||||
for _, entry := range entries {
|
||||
models = append(models, &ModelInfo{
|
||||
ID: entry.ID,
|
||||
Object: "model",
|
||||
Created: entry.Created,
|
||||
OwnedBy: "iflow",
|
||||
Type: "iflow",
|
||||
DisplayName: entry.DisplayName,
|
||||
Description: entry.Description,
|
||||
Thinking: entry.Thinking,
|
||||
// GetStaticModelDefinitionsByChannel returns static model definitions for a given channel/provider.
|
||||
// It returns nil when the channel is unknown.
|
||||
//
|
||||
// Supported channels:
|
||||
// - claude
|
||||
// - gemini
|
||||
// - vertex
|
||||
// - gemini-cli
|
||||
// - aistudio
|
||||
// - codex
|
||||
// - qwen
|
||||
// - iflow
|
||||
// - antigravity (returns static overrides only)
|
||||
func GetStaticModelDefinitionsByChannel(channel string) []*ModelInfo {
|
||||
key := strings.ToLower(strings.TrimSpace(channel))
|
||||
switch key {
|
||||
case "claude":
|
||||
return GetClaudeModels()
|
||||
case "gemini":
|
||||
return GetGeminiModels()
|
||||
case "vertex":
|
||||
return GetGeminiVertexModels()
|
||||
case "gemini-cli":
|
||||
return GetGeminiCLIModels()
|
||||
case "aistudio":
|
||||
return GetAIStudioModels()
|
||||
case "codex":
|
||||
return GetOpenAIModels()
|
||||
case "qwen":
|
||||
return GetQwenModels()
|
||||
case "iflow":
|
||||
return GetIFlowModels()
|
||||
case "antigravity":
|
||||
cfg := GetAntigravityModelConfig()
|
||||
if len(cfg) == 0 {
|
||||
return nil
|
||||
}
|
||||
models := make([]*ModelInfo, 0, len(cfg))
|
||||
for modelID, entry := range cfg {
|
||||
if modelID == "" || entry == nil {
|
||||
continue
|
||||
}
|
||||
models = append(models, &ModelInfo{
|
||||
ID: modelID,
|
||||
Object: "model",
|
||||
OwnedBy: "antigravity",
|
||||
Type: "antigravity",
|
||||
Thinking: entry.Thinking,
|
||||
MaxCompletionTokens: entry.MaxCompletionTokens,
|
||||
})
|
||||
}
|
||||
sort.Slice(models, func(i, j int) bool {
|
||||
return strings.ToLower(models[i].ID) < strings.ToLower(models[j].ID)
|
||||
})
|
||||
}
|
||||
return models
|
||||
}
|
||||
|
||||
// AntigravityModelConfig captures static antigravity model overrides, including
|
||||
// Thinking budget limits and provider max completion tokens.
|
||||
type AntigravityModelConfig struct {
|
||||
Thinking *ThinkingSupport
|
||||
MaxCompletionTokens int
|
||||
}
|
||||
|
||||
// GetAntigravityModelConfig returns static configuration for antigravity models.
|
||||
// Keys use upstream model names returned by the Antigravity models endpoint.
|
||||
func GetAntigravityModelConfig() map[string]*AntigravityModelConfig {
|
||||
return map[string]*AntigravityModelConfig{
|
||||
"gemini-2.5-flash": {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
|
||||
"gemini-2.5-flash-lite": {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
|
||||
"rev19-uic3-1p": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}},
|
||||
"gemini-3-pro-high": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
|
||||
"gemini-3-pro-image": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
|
||||
"gemini-3-flash": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}},
|
||||
"claude-sonnet-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
|
||||
"claude-opus-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
|
||||
"claude-sonnet-4-5": {MaxCompletionTokens: 64000},
|
||||
"gpt-oss-120b-medium": {},
|
||||
"tab_flash_lite_preview": {},
|
||||
return models
|
||||
default:
|
||||
return nil
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
846
internal/registry/model_definitions_static_data.go
Normal file
846
internal/registry/model_definitions_static_data.go
Normal file
@@ -0,0 +1,846 @@
|
||||
// Package registry provides model definitions for various AI service providers.
|
||||
// This file stores the static model metadata catalog.
|
||||
package registry
|
||||
|
||||
// GetClaudeModels returns the standard Claude model definitions
|
||||
func GetClaudeModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
|
||||
{
|
||||
ID: "claude-haiku-4-5-20251001",
|
||||
Object: "model",
|
||||
Created: 1759276800, // 2025-10-01
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Haiku",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
// Thinking: not supported for Haiku models
|
||||
},
|
||||
{
|
||||
ID: "claude-sonnet-4-5-20250929",
|
||||
Object: "model",
|
||||
Created: 1759104000, // 2025-09-29
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Sonnet",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-5-20251101",
|
||||
Object: "model",
|
||||
Created: 1761955200, // 2025-11-01
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.5 Opus",
|
||||
Description: "Premium model combining maximum intelligence with practical performance",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-1-20250805",
|
||||
Object: "model",
|
||||
Created: 1722945600, // 2025-08-05
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4.1 Opus",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 32000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-opus-4-20250514",
|
||||
Object: "model",
|
||||
Created: 1715644800, // 2025-05-14
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4 Opus",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 32000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-sonnet-4-20250514",
|
||||
Object: "model",
|
||||
Created: 1715644800, // 2025-05-14
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 4 Sonnet",
|
||||
ContextLength: 200000,
|
||||
MaxCompletionTokens: 64000,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-3-7-sonnet-20250219",
|
||||
Object: "model",
|
||||
Created: 1708300800, // 2025-02-19
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 3.7 Sonnet",
|
||||
ContextLength: 128000,
|
||||
MaxCompletionTokens: 8192,
|
||||
Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
|
||||
},
|
||||
{
|
||||
ID: "claude-3-5-haiku-20241022",
|
||||
Object: "model",
|
||||
Created: 1729555200, // 2024-10-22
|
||||
OwnedBy: "anthropic",
|
||||
Type: "claude",
|
||||
DisplayName: "Claude 3.5 Haiku",
|
||||
ContextLength: 128000,
|
||||
MaxCompletionTokens: 8192,
|
||||
// Thinking: not supported for Haiku models
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetGeminiModels returns the standard Gemini model definitions
|
||||
func GetGeminiModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Gemini 3 Flash Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-image-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-image-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Image Preview",
|
||||
Description: "Gemini 3 Pro Image Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
func GetGeminiVertexModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-image-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-image-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Image Preview",
|
||||
Description: "Gemini 3 Pro Image Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
// Imagen image generation models - use :predict action
|
||||
{
|
||||
ID: "imagen-4.0-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Generate",
|
||||
Description: "Imagen 4.0 image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-4.0-ultra-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-ultra-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Ultra Generate",
|
||||
Description: "Imagen 4.0 Ultra high-quality image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-3.0-generate-002",
|
||||
Object: "model",
|
||||
Created: 1740000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-3.0-generate-002",
|
||||
Version: "3.0",
|
||||
DisplayName: "Imagen 3.0 Generate",
|
||||
Description: "Imagen 3.0 image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-3.0-fast-generate-001",
|
||||
Object: "model",
|
||||
Created: 1740000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-3.0-fast-generate-001",
|
||||
Version: "3.0",
|
||||
DisplayName: "Imagen 3.0 Fast Generate",
|
||||
Description: "Imagen 3.0 fast image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
{
|
||||
ID: "imagen-4.0-fast-generate-001",
|
||||
Object: "model",
|
||||
Created: 1750000000,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/imagen-4.0-fast-generate-001",
|
||||
Version: "4.0",
|
||||
DisplayName: "Imagen 4.0 Fast Generate",
|
||||
Description: "Imagen 4.0 fast image generation model",
|
||||
SupportedGenerationMethods: []string{"predict"},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetGeminiCLIModels returns the standard Gemini model definitions
|
||||
func GetGeminiCLIModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Our most intelligent model with SOTA reasoning and multimodal understanding, and powerful agentic and vibe coding capabilities",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetAIStudioModels returns the Gemini model definitions for AI Studio integrations
|
||||
func GetAIStudioModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gemini-2.5-pro",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-pro",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Pro",
|
||||
Description: "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash",
|
||||
Version: "001",
|
||||
DisplayName: "Gemini 2.5 Flash",
|
||||
Description: "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-2.5-flash-lite",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-lite",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Lite",
|
||||
Description: "Our smallest and most cost effective model, built for at scale usage.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-pro-preview",
|
||||
Object: "model",
|
||||
Created: 1737158400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-pro-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Pro Preview",
|
||||
Description: "Gemini 3 Pro Preview",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-3-flash-preview",
|
||||
Object: "model",
|
||||
Created: 1765929600,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-3-flash-preview",
|
||||
Version: "3.0",
|
||||
DisplayName: "Gemini 3 Flash Preview",
|
||||
Description: "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-pro-latest",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-pro-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Pro Latest",
|
||||
Description: "Latest release of Gemini Pro",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-flash-latest",
|
||||
Object: "model",
|
||||
Created: 1750118400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-flash-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Flash Latest",
|
||||
Description: "Latest release of Gemini Flash",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
{
|
||||
ID: "gemini-flash-lite-latest",
|
||||
Object: "model",
|
||||
Created: 1753142400,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-flash-lite-latest",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini Flash-Lite Latest",
|
||||
Description: "Latest release of Gemini Flash-Lite",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 65536,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
Thinking: &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
|
||||
},
|
||||
// {
|
||||
// ID: "gemini-2.5-flash-image-preview",
|
||||
// Object: "model",
|
||||
// Created: 1756166400,
|
||||
// OwnedBy: "google",
|
||||
// Type: "gemini",
|
||||
// Name: "models/gemini-2.5-flash-image-preview",
|
||||
// Version: "2.5",
|
||||
// DisplayName: "Gemini 2.5 Flash Image Preview",
|
||||
// Description: "State-of-the-art image generation and editing model.",
|
||||
// InputTokenLimit: 1048576,
|
||||
// OutputTokenLimit: 8192,
|
||||
// SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
// // image models don't support thinkingConfig; leave Thinking nil
|
||||
// },
|
||||
{
|
||||
ID: "gemini-2.5-flash-image",
|
||||
Object: "model",
|
||||
Created: 1759363200,
|
||||
OwnedBy: "google",
|
||||
Type: "gemini",
|
||||
Name: "models/gemini-2.5-flash-image",
|
||||
Version: "2.5",
|
||||
DisplayName: "Gemini 2.5 Flash Image",
|
||||
Description: "State-of-the-art image generation and editing model.",
|
||||
InputTokenLimit: 1048576,
|
||||
OutputTokenLimit: 8192,
|
||||
SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
|
||||
// image models don't support thinkingConfig; leave Thinking nil
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetOpenAIModels returns the standard OpenAI model definitions
|
||||
func GetOpenAIModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "gpt-5",
|
||||
Object: "model",
|
||||
Created: 1754524800,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-08-07",
|
||||
DisplayName: "GPT 5",
|
||||
Description: "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"minimal", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5-codex",
|
||||
Object: "model",
|
||||
Created: 1757894400,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-09-15",
|
||||
DisplayName: "GPT 5 Codex",
|
||||
Description: "Stable version of GPT 5 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5-codex-mini",
|
||||
Object: "model",
|
||||
Created: 1762473600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5-2025-11-07",
|
||||
DisplayName: "GPT 5 Codex Mini",
|
||||
Description: "Stable version of GPT 5 Codex Mini: cheaper, faster, but less capable version of GPT 5 Codex.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5",
|
||||
Description: "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"none", "low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5.1 Codex",
|
||||
Description: "Stable version of GPT 5.1 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex-mini",
|
||||
Object: "model",
|
||||
Created: 1762905600,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-2025-11-12",
|
||||
DisplayName: "GPT 5.1 Codex Mini",
|
||||
Description: "Stable version of GPT 5.1 Codex Mini: cheaper, faster, but less capable version of GPT 5.1 Codex.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.1-codex-max",
|
||||
Object: "model",
|
||||
Created: 1763424000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.1-max",
|
||||
DisplayName: "GPT 5.1 Codex Max",
|
||||
Description: "Stable version of GPT 5.1 Codex Max",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.2",
|
||||
Object: "model",
|
||||
Created: 1765440000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.2",
|
||||
DisplayName: "GPT 5.2",
|
||||
Description: "Stable version of GPT 5.2",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"none", "low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
{
|
||||
ID: "gpt-5.2-codex",
|
||||
Object: "model",
|
||||
Created: 1765440000,
|
||||
OwnedBy: "openai",
|
||||
Type: "openai",
|
||||
Version: "gpt-5.2",
|
||||
DisplayName: "GPT 5.2 Codex",
|
||||
Description: "Stable version of GPT 5.2 Codex, The best model for coding and agentic tasks across domains.",
|
||||
ContextLength: 400000,
|
||||
MaxCompletionTokens: 128000,
|
||||
SupportedParameters: []string{"tools"},
|
||||
Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// GetQwenModels returns the standard Qwen model definitions
|
||||
func GetQwenModels() []*ModelInfo {
|
||||
return []*ModelInfo{
|
||||
{
|
||||
ID: "qwen3-coder-plus",
|
||||
Object: "model",
|
||||
Created: 1753228800,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Coder Plus",
|
||||
Description: "Advanced code generation and understanding model",
|
||||
ContextLength: 32768,
|
||||
MaxCompletionTokens: 8192,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
{
|
||||
ID: "qwen3-coder-flash",
|
||||
Object: "model",
|
||||
Created: 1753228800,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Coder Flash",
|
||||
Description: "Fast code generation model",
|
||||
ContextLength: 8192,
|
||||
MaxCompletionTokens: 2048,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
{
|
||||
ID: "vision-model",
|
||||
Object: "model",
|
||||
Created: 1758672000,
|
||||
OwnedBy: "qwen",
|
||||
Type: "qwen",
|
||||
Version: "3.0",
|
||||
DisplayName: "Qwen3 Vision Model",
|
||||
Description: "Vision model model",
|
||||
ContextLength: 32768,
|
||||
MaxCompletionTokens: 2048,
|
||||
SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
// iFlowThinkingSupport is a shared ThinkingSupport configuration for iFlow models
|
||||
// that support thinking mode via chat_template_kwargs.enable_thinking (boolean toggle).
|
||||
// Uses level-based configuration so standard normalization flows apply before conversion.
|
||||
var iFlowThinkingSupport = &ThinkingSupport{
|
||||
Levels: []string{"none", "auto", "minimal", "low", "medium", "high", "xhigh"},
|
||||
}
|
||||
|
||||
// GetIFlowModels returns supported models for iFlow OAuth accounts.
|
||||
func GetIFlowModels() []*ModelInfo {
|
||||
entries := []struct {
|
||||
ID string
|
||||
DisplayName string
|
||||
Description string
|
||||
Created int64
|
||||
Thinking *ThinkingSupport
|
||||
}{
|
||||
{ID: "tstars2.0", DisplayName: "TStars-2.0", Description: "iFlow TStars-2.0 multimodal assistant", Created: 1746489600},
|
||||
{ID: "qwen3-coder-plus", DisplayName: "Qwen3-Coder-Plus", Description: "Qwen3 Coder Plus code generation", Created: 1753228800},
|
||||
{ID: "qwen3-max", DisplayName: "Qwen3-Max", Description: "Qwen3 flagship model", Created: 1758672000},
|
||||
{ID: "qwen3-vl-plus", DisplayName: "Qwen3-VL-Plus", Description: "Qwen3 multimodal vision-language", Created: 1758672000},
|
||||
{ID: "qwen3-max-preview", DisplayName: "Qwen3-Max-Preview", Description: "Qwen3 Max preview build", Created: 1757030400, Thinking: iFlowThinkingSupport},
|
||||
{ID: "kimi-k2-0905", DisplayName: "Kimi-K2-Instruct-0905", Description: "Moonshot Kimi K2 instruct 0905", Created: 1757030400},
|
||||
{ID: "glm-4.6", DisplayName: "GLM-4.6", Description: "Zhipu GLM 4.6 general model", Created: 1759190400, Thinking: iFlowThinkingSupport},
|
||||
{ID: "glm-4.7", DisplayName: "GLM-4.7", Description: "Zhipu GLM 4.7 general model", Created: 1766448000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "kimi-k2", DisplayName: "Kimi-K2", Description: "Moonshot Kimi K2 general model", Created: 1752192000},
|
||||
{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 thinking model", Created: 1762387200},
|
||||
{ID: "deepseek-v3.2-chat", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Chat", Created: 1764576000},
|
||||
{ID: "deepseek-v3.2-reasoner", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Reasoner", Created: 1764576000},
|
||||
{ID: "deepseek-v3.2", DisplayName: "DeepSeek-V3.2-Exp", Description: "DeepSeek V3.2 experimental", Created: 1759104000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "deepseek-v3.1", DisplayName: "DeepSeek-V3.1-Terminus", Description: "DeepSeek V3.1 Terminus", Created: 1756339200, Thinking: iFlowThinkingSupport},
|
||||
{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200},
|
||||
{ID: "deepseek-v3", DisplayName: "DeepSeek-V3-671B", Description: "DeepSeek V3 671B", Created: 1734307200},
|
||||
{ID: "qwen3-32b", DisplayName: "Qwen3-32B", Description: "Qwen3 32B", Created: 1747094400},
|
||||
{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600},
|
||||
{ID: "qwen3-235b-a22b-instruct", DisplayName: "Qwen3-235B-A22B-Instruct", Description: "Qwen3 235B A22B Instruct", Created: 1753401600},
|
||||
{ID: "qwen3-235b", DisplayName: "Qwen3-235B-A22B", Description: "Qwen3 235B A22B", Created: 1753401600},
|
||||
{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "minimax-m2.1", DisplayName: "MiniMax-M2.1", Description: "MiniMax M2.1", Created: 1766448000, Thinking: iFlowThinkingSupport},
|
||||
{ID: "iflow-rome-30ba3b", DisplayName: "iFlow-ROME", Description: "iFlow Rome 30BA3B model", Created: 1736899200},
|
||||
}
|
||||
models := make([]*ModelInfo, 0, len(entries))
|
||||
for _, entry := range entries {
|
||||
models = append(models, &ModelInfo{
|
||||
ID: entry.ID,
|
||||
Object: "model",
|
||||
Created: entry.Created,
|
||||
OwnedBy: "iflow",
|
||||
Type: "iflow",
|
||||
DisplayName: entry.DisplayName,
|
||||
Description: entry.Description,
|
||||
Thinking: entry.Thinking,
|
||||
})
|
||||
}
|
||||
return models
|
||||
}
|
||||
|
||||
// AntigravityModelConfig captures static antigravity model overrides, including
|
||||
// Thinking budget limits and provider max completion tokens.
|
||||
type AntigravityModelConfig struct {
|
||||
Thinking *ThinkingSupport
|
||||
MaxCompletionTokens int
|
||||
}
|
||||
|
||||
// GetAntigravityModelConfig returns static configuration for antigravity models.
|
||||
// Keys use upstream model names returned by the Antigravity models endpoint.
|
||||
func GetAntigravityModelConfig() map[string]*AntigravityModelConfig {
|
||||
return map[string]*AntigravityModelConfig{
|
||||
// "rev19-uic3-1p": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}},
|
||||
"gemini-2.5-flash": {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
|
||||
"gemini-2.5-flash-lite": {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
|
||||
"gemini-3-pro-high": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
|
||||
"gemini-3-pro-image": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
|
||||
"gemini-3-flash": {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}},
|
||||
"claude-sonnet-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
|
||||
"claude-opus-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
|
||||
"claude-sonnet-4-5": {MaxCompletionTokens: 64000},
|
||||
"gpt-oss-120b-medium": {},
|
||||
"tab_flash_lite_preview": {},
|
||||
}
|
||||
}
|
||||
@@ -148,7 +148,7 @@ func (e *AntigravityExecutor) Execute(ctx context.Context, auth *cliproxyauth.Au
|
||||
baseURLs := antigravityBaseURLFallbackOrder(auth)
|
||||
httpClient := newProxyAwareHTTPClient(ctx, e.cfg, auth, 0)
|
||||
|
||||
attempts := antigravityRetryAttempts(e.cfg)
|
||||
attempts := antigravityRetryAttempts(auth, e.cfg)
|
||||
|
||||
attemptLoop:
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
@@ -289,7 +289,7 @@ func (e *AntigravityExecutor) executeClaudeNonStream(ctx context.Context, auth *
|
||||
baseURLs := antigravityBaseURLFallbackOrder(auth)
|
||||
httpClient := newProxyAwareHTTPClient(ctx, e.cfg, auth, 0)
|
||||
|
||||
attempts := antigravityRetryAttempts(e.cfg)
|
||||
attempts := antigravityRetryAttempts(auth, e.cfg)
|
||||
|
||||
attemptLoop:
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
@@ -677,7 +677,7 @@ func (e *AntigravityExecutor) ExecuteStream(ctx context.Context, auth *cliproxya
|
||||
baseURLs := antigravityBaseURLFallbackOrder(auth)
|
||||
httpClient := newProxyAwareHTTPClient(ctx, e.cfg, auth, 0)
|
||||
|
||||
attempts := antigravityRetryAttempts(e.cfg)
|
||||
attempts := antigravityRetryAttempts(auth, e.cfg)
|
||||
|
||||
attemptLoop:
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
@@ -1447,11 +1447,16 @@ func resolveUserAgent(auth *cliproxyauth.Auth) string {
|
||||
return defaultAntigravityAgent
|
||||
}
|
||||
|
||||
func antigravityRetryAttempts(cfg *config.Config) int {
|
||||
if cfg == nil {
|
||||
return 1
|
||||
func antigravityRetryAttempts(auth *cliproxyauth.Auth, cfg *config.Config) int {
|
||||
retry := 0
|
||||
if cfg != nil {
|
||||
retry = cfg.RequestRetry
|
||||
}
|
||||
if auth != nil {
|
||||
if override, ok := auth.RequestRetryOverride(); ok {
|
||||
retry = override
|
||||
}
|
||||
}
|
||||
retry := cfg.RequestRetry
|
||||
if retry < 0 {
|
||||
retry = 0
|
||||
}
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
// Package iflow implements thinking configuration for iFlow models (GLM, MiniMax).
|
||||
// Package iflow implements thinking configuration for iFlow models.
|
||||
//
|
||||
// iFlow models use boolean toggle semantics:
|
||||
// - GLM models: chat_template_kwargs.enable_thinking (boolean)
|
||||
// - Models using chat_template_kwargs.enable_thinking (boolean toggle)
|
||||
// - MiniMax models: reasoning_split (boolean)
|
||||
//
|
||||
// Level values are converted to boolean: none=false, all others=true
|
||||
@@ -20,6 +20,7 @@ import (
|
||||
// Applier implements thinking.ProviderApplier for iFlow models.
|
||||
//
|
||||
// iFlow-specific behavior:
|
||||
// - enable_thinking toggle models: enable_thinking boolean
|
||||
// - GLM models: enable_thinking boolean + clear_thinking=false
|
||||
// - MiniMax models: reasoning_split boolean
|
||||
// - Level to boolean: none=false, others=true
|
||||
@@ -61,8 +62,8 @@ func (a *Applier) Apply(body []byte, config thinking.ThinkingConfig, modelInfo *
|
||||
return body, nil
|
||||
}
|
||||
|
||||
if isGLMModel(modelInfo.ID) {
|
||||
return applyGLM(body, config), nil
|
||||
if isEnableThinkingModel(modelInfo.ID) {
|
||||
return applyEnableThinking(body, config, isGLMModel(modelInfo.ID)), nil
|
||||
}
|
||||
|
||||
if isMiniMaxModel(modelInfo.ID) {
|
||||
@@ -97,7 +98,8 @@ func configToBoolean(config thinking.ThinkingConfig) bool {
|
||||
}
|
||||
}
|
||||
|
||||
// applyGLM applies thinking configuration for GLM models.
|
||||
// applyEnableThinking applies thinking configuration for models that use
|
||||
// chat_template_kwargs.enable_thinking format.
|
||||
//
|
||||
// Output format when enabled:
|
||||
//
|
||||
@@ -107,9 +109,8 @@ func configToBoolean(config thinking.ThinkingConfig) bool {
|
||||
//
|
||||
// {"chat_template_kwargs": {"enable_thinking": false}}
|
||||
//
|
||||
// Note: clear_thinking is only set when thinking is enabled, to preserve
|
||||
// thinking output in the response.
|
||||
func applyGLM(body []byte, config thinking.ThinkingConfig) []byte {
|
||||
// Note: clear_thinking is only set for GLM models when thinking is enabled.
|
||||
func applyEnableThinking(body []byte, config thinking.ThinkingConfig, setClearThinking bool) []byte {
|
||||
enableThinking := configToBoolean(config)
|
||||
|
||||
if len(body) == 0 || !gjson.ValidBytes(body) {
|
||||
@@ -118,8 +119,11 @@ func applyGLM(body []byte, config thinking.ThinkingConfig) []byte {
|
||||
|
||||
result, _ := sjson.SetBytes(body, "chat_template_kwargs.enable_thinking", enableThinking)
|
||||
|
||||
// clear_thinking is a GLM-only knob, strip it for other models.
|
||||
result, _ = sjson.DeleteBytes(result, "chat_template_kwargs.clear_thinking")
|
||||
|
||||
// clear_thinking only needed when thinking is enabled
|
||||
if enableThinking {
|
||||
if enableThinking && setClearThinking {
|
||||
result, _ = sjson.SetBytes(result, "chat_template_kwargs.clear_thinking", false)
|
||||
}
|
||||
|
||||
@@ -143,8 +147,21 @@ func applyMiniMax(body []byte, config thinking.ThinkingConfig) []byte {
|
||||
return result
|
||||
}
|
||||
|
||||
// isEnableThinkingModel determines if the model uses chat_template_kwargs.enable_thinking format.
|
||||
func isEnableThinkingModel(modelID string) bool {
|
||||
if isGLMModel(modelID) {
|
||||
return true
|
||||
}
|
||||
id := strings.ToLower(modelID)
|
||||
switch id {
|
||||
case "qwen3-max-preview", "deepseek-v3.2", "deepseek-v3.1":
|
||||
return true
|
||||
default:
|
||||
return false
|
||||
}
|
||||
}
|
||||
|
||||
// isGLMModel determines if the model is a GLM series model.
|
||||
// GLM models use chat_template_kwargs.enable_thinking format.
|
||||
func isGLMModel(modelID string) bool {
|
||||
return strings.HasPrefix(strings.ToLower(modelID), "glm")
|
||||
}
|
||||
|
||||
@@ -175,7 +175,7 @@ func convertConstToEnum(jsonStr string) string {
|
||||
return jsonStr
|
||||
}
|
||||
|
||||
// convertEnumValuesToStrings ensures all enum values are strings.
|
||||
// convertEnumValuesToStrings ensures all enum values are strings and the schema type is set to string.
|
||||
// Gemini API requires enum values to be of type string, not numbers or booleans.
|
||||
func convertEnumValuesToStrings(jsonStr string) string {
|
||||
for _, p := range findPaths(jsonStr, "enum") {
|
||||
@@ -185,19 +185,15 @@ func convertEnumValuesToStrings(jsonStr string) string {
|
||||
}
|
||||
|
||||
var stringVals []string
|
||||
needsConversion := false
|
||||
for _, item := range arr.Array() {
|
||||
// Check if any value is not a string
|
||||
if item.Type != gjson.String {
|
||||
needsConversion = true
|
||||
}
|
||||
stringVals = append(stringVals, item.String())
|
||||
}
|
||||
|
||||
// Only update if we found non-string values
|
||||
if needsConversion {
|
||||
jsonStr, _ = sjson.Set(jsonStr, p, stringVals)
|
||||
}
|
||||
// Always update enum values to strings and set type to "string"
|
||||
// This ensures compatibility with Antigravity Gemini which only allows enum for STRING type
|
||||
jsonStr, _ = sjson.Set(jsonStr, p, stringVals)
|
||||
parentPath := trimSuffix(p, ".enum")
|
||||
jsonStr, _ = sjson.Set(jsonStr, joinPath(parentPath, "type"), "string")
|
||||
}
|
||||
return jsonStr
|
||||
}
|
||||
|
||||
@@ -86,12 +86,19 @@ func (s *FileSynthesizer) Synthesize(ctx *SynthesisContext) ([]*coreauth.Auth, e
|
||||
}
|
||||
}
|
||||
|
||||
disabled, _ := metadata["disabled"].(bool)
|
||||
status := coreauth.StatusActive
|
||||
if disabled {
|
||||
status = coreauth.StatusDisabled
|
||||
}
|
||||
|
||||
a := &coreauth.Auth{
|
||||
ID: id,
|
||||
Provider: provider,
|
||||
Label: label,
|
||||
Prefix: prefix,
|
||||
Status: coreauth.StatusActive,
|
||||
Status: status,
|
||||
Disabled: disabled,
|
||||
Attributes: map[string]string{
|
||||
"source": full,
|
||||
"path": full,
|
||||
@@ -167,6 +174,16 @@ func SynthesizeGeminiVirtualAuths(primary *coreauth.Auth, metadata map[string]an
|
||||
"virtual_parent_id": primary.ID,
|
||||
"type": metadata["type"],
|
||||
}
|
||||
if v, ok := metadata["disable_cooling"]; ok {
|
||||
metadataCopy["disable_cooling"] = v
|
||||
} else if v, ok := metadata["disable-cooling"]; ok {
|
||||
metadataCopy["disable_cooling"] = v
|
||||
}
|
||||
if v, ok := metadata["request_retry"]; ok {
|
||||
metadataCopy["request_retry"] = v
|
||||
} else if v, ok := metadata["request-retry"]; ok {
|
||||
metadataCopy["request_retry"] = v
|
||||
}
|
||||
proxy := strings.TrimSpace(primary.ProxyURL)
|
||||
if proxy != "" {
|
||||
metadataCopy["proxy_url"] = proxy
|
||||
|
||||
@@ -69,10 +69,12 @@ func TestFileSynthesizer_Synthesize_ValidAuthFile(t *testing.T) {
|
||||
|
||||
// Create a valid auth file
|
||||
authData := map[string]any{
|
||||
"type": "claude",
|
||||
"email": "test@example.com",
|
||||
"proxy_url": "http://proxy.local",
|
||||
"prefix": "test-prefix",
|
||||
"type": "claude",
|
||||
"email": "test@example.com",
|
||||
"proxy_url": "http://proxy.local",
|
||||
"prefix": "test-prefix",
|
||||
"disable_cooling": true,
|
||||
"request_retry": 2,
|
||||
}
|
||||
data, _ := json.Marshal(authData)
|
||||
err := os.WriteFile(filepath.Join(tempDir, "claude-auth.json"), data, 0644)
|
||||
@@ -108,6 +110,12 @@ func TestFileSynthesizer_Synthesize_ValidAuthFile(t *testing.T) {
|
||||
if auths[0].ProxyURL != "http://proxy.local" {
|
||||
t.Errorf("expected proxy_url http://proxy.local, got %s", auths[0].ProxyURL)
|
||||
}
|
||||
if v, ok := auths[0].Metadata["disable_cooling"].(bool); !ok || !v {
|
||||
t.Errorf("expected disable_cooling true, got %v", auths[0].Metadata["disable_cooling"])
|
||||
}
|
||||
if v, ok := auths[0].Metadata["request_retry"].(float64); !ok || int(v) != 2 {
|
||||
t.Errorf("expected request_retry 2, got %v", auths[0].Metadata["request_retry"])
|
||||
}
|
||||
if auths[0].Status != coreauth.StatusActive {
|
||||
t.Errorf("expected status active, got %s", auths[0].Status)
|
||||
}
|
||||
@@ -336,9 +344,11 @@ func TestSynthesizeGeminiVirtualAuths_MultiProject(t *testing.T) {
|
||||
},
|
||||
}
|
||||
metadata := map[string]any{
|
||||
"project_id": "project-a, project-b, project-c",
|
||||
"email": "test@example.com",
|
||||
"type": "gemini",
|
||||
"project_id": "project-a, project-b, project-c",
|
||||
"email": "test@example.com",
|
||||
"type": "gemini",
|
||||
"request_retry": 2,
|
||||
"disable_cooling": true,
|
||||
}
|
||||
|
||||
virtuals := SynthesizeGeminiVirtualAuths(primary, metadata, now)
|
||||
@@ -376,6 +386,12 @@ func TestSynthesizeGeminiVirtualAuths_MultiProject(t *testing.T) {
|
||||
if v.ProxyURL != "http://proxy.local" {
|
||||
t.Errorf("expected proxy_url http://proxy.local, got %s", v.ProxyURL)
|
||||
}
|
||||
if vv, ok := v.Metadata["disable_cooling"].(bool); !ok || !vv {
|
||||
t.Errorf("expected disable_cooling true, got %v", v.Metadata["disable_cooling"])
|
||||
}
|
||||
if vv, ok := v.Metadata["request_retry"].(int); !ok || vv != 2 {
|
||||
t.Errorf("expected request_retry 2, got %v", v.Metadata["request_retry"])
|
||||
}
|
||||
if v.Attributes["runtime_only"] != "true" {
|
||||
t.Error("expected runtime_only=true")
|
||||
}
|
||||
|
||||
@@ -124,32 +124,47 @@ func (m *Manager) Stream(ctx context.Context, provider string, req *HTTPRequest)
|
||||
out := make(chan StreamEvent)
|
||||
go func() {
|
||||
defer close(out)
|
||||
send := func(ev StreamEvent) bool {
|
||||
if ctx == nil {
|
||||
out <- ev
|
||||
return true
|
||||
}
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
return false
|
||||
case out <- ev:
|
||||
return true
|
||||
}
|
||||
}
|
||||
for {
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
out <- StreamEvent{Err: ctx.Err()}
|
||||
return
|
||||
case msg, ok := <-respCh:
|
||||
if !ok {
|
||||
out <- StreamEvent{Err: errors.New("wsrelay: stream closed")}
|
||||
_ = send(StreamEvent{Err: errors.New("wsrelay: stream closed")})
|
||||
return
|
||||
}
|
||||
switch msg.Type {
|
||||
case MessageTypeStreamStart:
|
||||
resp := decodeResponse(msg.Payload)
|
||||
out <- StreamEvent{Type: MessageTypeStreamStart, Status: resp.Status, Headers: resp.Headers}
|
||||
if okSend := send(StreamEvent{Type: MessageTypeStreamStart, Status: resp.Status, Headers: resp.Headers}); !okSend {
|
||||
return
|
||||
}
|
||||
case MessageTypeStreamChunk:
|
||||
chunk := decodeChunk(msg.Payload)
|
||||
out <- StreamEvent{Type: MessageTypeStreamChunk, Payload: chunk}
|
||||
if okSend := send(StreamEvent{Type: MessageTypeStreamChunk, Payload: chunk}); !okSend {
|
||||
return
|
||||
}
|
||||
case MessageTypeStreamEnd:
|
||||
out <- StreamEvent{Type: MessageTypeStreamEnd}
|
||||
_ = send(StreamEvent{Type: MessageTypeStreamEnd})
|
||||
return
|
||||
case MessageTypeError:
|
||||
out <- StreamEvent{Type: MessageTypeError, Err: decodeError(msg.Payload)}
|
||||
_ = send(StreamEvent{Type: MessageTypeError, Err: decodeError(msg.Payload)})
|
||||
return
|
||||
case MessageTypeHTTPResp:
|
||||
resp := decodeResponse(msg.Payload)
|
||||
out <- StreamEvent{Type: MessageTypeHTTPResp, Status: resp.Status, Headers: resp.Headers, Payload: resp.Body}
|
||||
_ = send(StreamEvent{Type: MessageTypeHTTPResp, Status: resp.Status, Headers: resp.Headers, Payload: resp.Body})
|
||||
return
|
||||
default:
|
||||
}
|
||||
|
||||
@@ -506,6 +506,32 @@ func (h *BaseAPIHandler) ExecuteStreamWithAuthManager(ctx context.Context, handl
|
||||
bootstrapRetries := 0
|
||||
maxBootstrapRetries := StreamingBootstrapRetries(h.Cfg)
|
||||
|
||||
sendErr := func(msg *interfaces.ErrorMessage) bool {
|
||||
if ctx == nil {
|
||||
errChan <- msg
|
||||
return true
|
||||
}
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
return false
|
||||
case errChan <- msg:
|
||||
return true
|
||||
}
|
||||
}
|
||||
|
||||
sendData := func(chunk []byte) bool {
|
||||
if ctx == nil {
|
||||
dataChan <- chunk
|
||||
return true
|
||||
}
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
return false
|
||||
case dataChan <- chunk:
|
||||
return true
|
||||
}
|
||||
}
|
||||
|
||||
bootstrapEligible := func(err error) bool {
|
||||
status := statusFromError(err)
|
||||
if status == 0 {
|
||||
@@ -565,12 +591,14 @@ func (h *BaseAPIHandler) ExecuteStreamWithAuthManager(ctx context.Context, handl
|
||||
addon = hdr.Clone()
|
||||
}
|
||||
}
|
||||
errChan <- &interfaces.ErrorMessage{StatusCode: status, Error: streamErr, Addon: addon}
|
||||
_ = sendErr(&interfaces.ErrorMessage{StatusCode: status, Error: streamErr, Addon: addon})
|
||||
return
|
||||
}
|
||||
if len(chunk.Payload) > 0 {
|
||||
sentPayload = true
|
||||
dataChan <- cloneBytes(chunk.Payload)
|
||||
if okSendData := sendData(cloneBytes(chunk.Payload)); !okSendData {
|
||||
return
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -68,14 +68,13 @@ func (s *FileTokenStore) Save(ctx context.Context, auth *cliproxyauth.Auth) (str
|
||||
return "", err
|
||||
}
|
||||
case auth.Metadata != nil:
|
||||
auth.Metadata["disabled"] = auth.Disabled
|
||||
raw, errMarshal := json.Marshal(auth.Metadata)
|
||||
if errMarshal != nil {
|
||||
return "", fmt.Errorf("auth filestore: marshal metadata failed: %w", errMarshal)
|
||||
}
|
||||
if existing, errRead := os.ReadFile(path); errRead == nil {
|
||||
// Use metadataEqualIgnoringTimestamps to skip writes when only timestamp fields change.
|
||||
// This prevents the token refresh loop caused by timestamp/expired/expires_in changes.
|
||||
if metadataEqualIgnoringTimestamps(existing, raw, auth.Provider) {
|
||||
if jsonEqual(existing, raw) {
|
||||
return path, nil
|
||||
}
|
||||
file, errOpen := os.OpenFile(path, os.O_WRONLY|os.O_TRUNC, 0o600)
|
||||
@@ -216,12 +215,18 @@ func (s *FileTokenStore) readAuthFile(path, baseDir string) (*cliproxyauth.Auth,
|
||||
return nil, fmt.Errorf("stat file: %w", err)
|
||||
}
|
||||
id := s.idFor(path, baseDir)
|
||||
disabled, _ := metadata["disabled"].(bool)
|
||||
status := cliproxyauth.StatusActive
|
||||
if disabled {
|
||||
status = cliproxyauth.StatusDisabled
|
||||
}
|
||||
auth := &cliproxyauth.Auth{
|
||||
ID: id,
|
||||
Provider: provider,
|
||||
FileName: id,
|
||||
Label: s.labelFor(metadata),
|
||||
Status: cliproxyauth.StatusActive,
|
||||
Status: status,
|
||||
Disabled: disabled,
|
||||
Attributes: map[string]string{"path": path},
|
||||
Metadata: metadata,
|
||||
CreatedAt: info.ModTime(),
|
||||
@@ -299,8 +304,7 @@ func (s *FileTokenStore) baseDirSnapshot() string {
|
||||
return s.baseDir
|
||||
}
|
||||
|
||||
// DEPRECATED: Use metadataEqualIgnoringTimestamps for comparing auth metadata.
|
||||
// This function is kept for backward compatibility but can cause refresh loops.
|
||||
// jsonEqual compares two JSON blobs by parsing them into Go objects and deep comparing.
|
||||
func jsonEqual(a, b []byte) bool {
|
||||
var objA any
|
||||
var objB any
|
||||
@@ -313,41 +317,6 @@ func jsonEqual(a, b []byte) bool {
|
||||
return deepEqualJSON(objA, objB)
|
||||
}
|
||||
|
||||
// metadataEqualIgnoringTimestamps compares two metadata JSON blobs,
|
||||
// ignoring fields that change on every refresh but don't affect functionality.
|
||||
// This prevents unnecessary file writes that would trigger watcher events and
|
||||
// create refresh loops.
|
||||
// The provider parameter controls whether access_token is ignored: providers like
|
||||
// Google OAuth (gemini, gemini-cli) can re-fetch tokens when needed, while others
|
||||
// like iFlow require the refreshed token to be persisted.
|
||||
func metadataEqualIgnoringTimestamps(a, b []byte, provider string) bool {
|
||||
var objA, objB map[string]any
|
||||
if err := json.Unmarshal(a, &objA); err != nil {
|
||||
return false
|
||||
}
|
||||
if err := json.Unmarshal(b, &objB); err != nil {
|
||||
return false
|
||||
}
|
||||
|
||||
// Fields to ignore: these change on every refresh but don't affect authentication logic.
|
||||
// - timestamp, expired, expires_in, last_refresh: time-related fields that change on refresh
|
||||
ignoredFields := []string{"timestamp", "expired", "expires_in", "last_refresh"}
|
||||
|
||||
// For providers that can re-fetch tokens when needed (e.g., Google OAuth),
|
||||
// we ignore access_token to avoid unnecessary file writes.
|
||||
switch provider {
|
||||
case "gemini", "gemini-cli", "antigravity":
|
||||
ignoredFields = append(ignoredFields, "access_token")
|
||||
}
|
||||
|
||||
for _, field := range ignoredFields {
|
||||
delete(objA, field)
|
||||
delete(objB, field)
|
||||
}
|
||||
|
||||
return deepEqualJSON(objA, objB)
|
||||
}
|
||||
|
||||
func deepEqualJSON(a, b any) bool {
|
||||
switch valA := a.(type) {
|
||||
case map[string]any:
|
||||
|
||||
@@ -61,6 +61,15 @@ func SetQuotaCooldownDisabled(disable bool) {
|
||||
quotaCooldownDisabled.Store(disable)
|
||||
}
|
||||
|
||||
func quotaCooldownDisabledForAuth(auth *Auth) bool {
|
||||
if auth != nil {
|
||||
if override, ok := auth.DisableCoolingOverride(); ok {
|
||||
return override
|
||||
}
|
||||
}
|
||||
return quotaCooldownDisabled.Load()
|
||||
}
|
||||
|
||||
// Result captures execution outcome used to adjust auth state.
|
||||
type Result struct {
|
||||
// AuthID references the auth that produced this result.
|
||||
@@ -468,20 +477,16 @@ func (m *Manager) Execute(ctx context.Context, providers []string, req cliproxye
|
||||
return cliproxyexecutor.Response{}, &Error{Code: "provider_not_found", Message: "no provider supplied"}
|
||||
}
|
||||
|
||||
retryTimes, maxWait := m.retrySettings()
|
||||
attempts := retryTimes + 1
|
||||
if attempts < 1 {
|
||||
attempts = 1
|
||||
}
|
||||
_, maxWait := m.retrySettings()
|
||||
|
||||
var lastErr error
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
for attempt := 0; ; attempt++ {
|
||||
resp, errExec := m.executeMixedOnce(ctx, normalized, req, opts)
|
||||
if errExec == nil {
|
||||
return resp, nil
|
||||
}
|
||||
lastErr = errExec
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errExec, attempt, attempts, normalized, req.Model, maxWait)
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errExec, attempt, normalized, req.Model, maxWait)
|
||||
if !shouldRetry {
|
||||
break
|
||||
}
|
||||
@@ -503,20 +508,16 @@ func (m *Manager) ExecuteCount(ctx context.Context, providers []string, req clip
|
||||
return cliproxyexecutor.Response{}, &Error{Code: "provider_not_found", Message: "no provider supplied"}
|
||||
}
|
||||
|
||||
retryTimes, maxWait := m.retrySettings()
|
||||
attempts := retryTimes + 1
|
||||
if attempts < 1 {
|
||||
attempts = 1
|
||||
}
|
||||
_, maxWait := m.retrySettings()
|
||||
|
||||
var lastErr error
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
for attempt := 0; ; attempt++ {
|
||||
resp, errExec := m.executeCountMixedOnce(ctx, normalized, req, opts)
|
||||
if errExec == nil {
|
||||
return resp, nil
|
||||
}
|
||||
lastErr = errExec
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errExec, attempt, attempts, normalized, req.Model, maxWait)
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errExec, attempt, normalized, req.Model, maxWait)
|
||||
if !shouldRetry {
|
||||
break
|
||||
}
|
||||
@@ -538,20 +539,16 @@ func (m *Manager) ExecuteStream(ctx context.Context, providers []string, req cli
|
||||
return nil, &Error{Code: "provider_not_found", Message: "no provider supplied"}
|
||||
}
|
||||
|
||||
retryTimes, maxWait := m.retrySettings()
|
||||
attempts := retryTimes + 1
|
||||
if attempts < 1 {
|
||||
attempts = 1
|
||||
}
|
||||
_, maxWait := m.retrySettings()
|
||||
|
||||
var lastErr error
|
||||
for attempt := 0; attempt < attempts; attempt++ {
|
||||
for attempt := 0; ; attempt++ {
|
||||
chunks, errStream := m.executeStreamMixedOnce(ctx, normalized, req, opts)
|
||||
if errStream == nil {
|
||||
return chunks, nil
|
||||
}
|
||||
lastErr = errStream
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errStream, attempt, attempts, normalized, req.Model, maxWait)
|
||||
wait, shouldRetry := m.shouldRetryAfterError(errStream, attempt, normalized, req.Model, maxWait)
|
||||
if !shouldRetry {
|
||||
break
|
||||
}
|
||||
@@ -721,6 +718,7 @@ func (m *Manager) executeStreamMixedOnce(ctx context.Context, providers []string
|
||||
go func(streamCtx context.Context, streamAuth *Auth, streamProvider string, streamChunks <-chan cliproxyexecutor.StreamChunk) {
|
||||
defer close(out)
|
||||
var failed bool
|
||||
forward := true
|
||||
for chunk := range streamChunks {
|
||||
if chunk.Err != nil && !failed {
|
||||
failed = true
|
||||
@@ -731,7 +729,18 @@ func (m *Manager) executeStreamMixedOnce(ctx context.Context, providers []string
|
||||
}
|
||||
m.MarkResult(streamCtx, Result{AuthID: streamAuth.ID, Provider: streamProvider, Model: routeModel, Success: false, Error: rerr})
|
||||
}
|
||||
out <- chunk
|
||||
if !forward {
|
||||
continue
|
||||
}
|
||||
if streamCtx == nil {
|
||||
out <- chunk
|
||||
continue
|
||||
}
|
||||
select {
|
||||
case <-streamCtx.Done():
|
||||
forward = false
|
||||
case out <- chunk:
|
||||
}
|
||||
}
|
||||
if !failed {
|
||||
m.MarkResult(streamCtx, Result{AuthID: streamAuth.ID, Provider: streamProvider, Model: routeModel, Success: true})
|
||||
@@ -1034,11 +1043,15 @@ func (m *Manager) retrySettings() (int, time.Duration) {
|
||||
return int(m.requestRetry.Load()), time.Duration(m.maxRetryInterval.Load())
|
||||
}
|
||||
|
||||
func (m *Manager) closestCooldownWait(providers []string, model string) (time.Duration, bool) {
|
||||
func (m *Manager) closestCooldownWait(providers []string, model string, attempt int) (time.Duration, bool) {
|
||||
if m == nil || len(providers) == 0 {
|
||||
return 0, false
|
||||
}
|
||||
now := time.Now()
|
||||
defaultRetry := int(m.requestRetry.Load())
|
||||
if defaultRetry < 0 {
|
||||
defaultRetry = 0
|
||||
}
|
||||
providerSet := make(map[string]struct{}, len(providers))
|
||||
for i := range providers {
|
||||
key := strings.TrimSpace(strings.ToLower(providers[i]))
|
||||
@@ -1061,6 +1074,16 @@ func (m *Manager) closestCooldownWait(providers []string, model string) (time.Du
|
||||
if _, ok := providerSet[providerKey]; !ok {
|
||||
continue
|
||||
}
|
||||
effectiveRetry := defaultRetry
|
||||
if override, ok := auth.RequestRetryOverride(); ok {
|
||||
effectiveRetry = override
|
||||
}
|
||||
if effectiveRetry < 0 {
|
||||
effectiveRetry = 0
|
||||
}
|
||||
if attempt >= effectiveRetry {
|
||||
continue
|
||||
}
|
||||
blocked, reason, next := isAuthBlockedForModel(auth, model, now)
|
||||
if !blocked || next.IsZero() || reason == blockReasonDisabled {
|
||||
continue
|
||||
@@ -1077,8 +1100,8 @@ func (m *Manager) closestCooldownWait(providers []string, model string) (time.Du
|
||||
return minWait, found
|
||||
}
|
||||
|
||||
func (m *Manager) shouldRetryAfterError(err error, attempt, maxAttempts int, providers []string, model string, maxWait time.Duration) (time.Duration, bool) {
|
||||
if err == nil || attempt >= maxAttempts-1 {
|
||||
func (m *Manager) shouldRetryAfterError(err error, attempt int, providers []string, model string, maxWait time.Duration) (time.Duration, bool) {
|
||||
if err == nil {
|
||||
return 0, false
|
||||
}
|
||||
if maxWait <= 0 {
|
||||
@@ -1087,7 +1110,7 @@ func (m *Manager) shouldRetryAfterError(err error, attempt, maxAttempts int, pro
|
||||
if status := statusCodeFromError(err); status == http.StatusOK {
|
||||
return 0, false
|
||||
}
|
||||
wait, found := m.closestCooldownWait(providers, model)
|
||||
wait, found := m.closestCooldownWait(providers, model, attempt)
|
||||
if !found || wait > maxWait {
|
||||
return 0, false
|
||||
}
|
||||
@@ -1176,7 +1199,7 @@ func (m *Manager) MarkResult(ctx context.Context, result Result) {
|
||||
if result.RetryAfter != nil {
|
||||
next = now.Add(*result.RetryAfter)
|
||||
} else {
|
||||
cooldown, nextLevel := nextQuotaCooldown(backoffLevel)
|
||||
cooldown, nextLevel := nextQuotaCooldown(backoffLevel, quotaCooldownDisabledForAuth(auth))
|
||||
if cooldown > 0 {
|
||||
next = now.Add(cooldown)
|
||||
}
|
||||
@@ -1193,7 +1216,7 @@ func (m *Manager) MarkResult(ctx context.Context, result Result) {
|
||||
shouldSuspendModel = true
|
||||
setModelQuota = true
|
||||
case 408, 500, 502, 503, 504:
|
||||
if quotaCooldownDisabled.Load() {
|
||||
if quotaCooldownDisabledForAuth(auth) {
|
||||
state.NextRetryAfter = time.Time{}
|
||||
} else {
|
||||
next := now.Add(1 * time.Minute)
|
||||
@@ -1439,7 +1462,7 @@ func applyAuthFailureState(auth *Auth, resultErr *Error, retryAfter *time.Durati
|
||||
if retryAfter != nil {
|
||||
next = now.Add(*retryAfter)
|
||||
} else {
|
||||
cooldown, nextLevel := nextQuotaCooldown(auth.Quota.BackoffLevel)
|
||||
cooldown, nextLevel := nextQuotaCooldown(auth.Quota.BackoffLevel, quotaCooldownDisabledForAuth(auth))
|
||||
if cooldown > 0 {
|
||||
next = now.Add(cooldown)
|
||||
}
|
||||
@@ -1449,7 +1472,7 @@ func applyAuthFailureState(auth *Auth, resultErr *Error, retryAfter *time.Durati
|
||||
auth.NextRetryAfter = next
|
||||
case 408, 500, 502, 503, 504:
|
||||
auth.StatusMessage = "transient upstream error"
|
||||
if quotaCooldownDisabled.Load() {
|
||||
if quotaCooldownDisabledForAuth(auth) {
|
||||
auth.NextRetryAfter = time.Time{}
|
||||
} else {
|
||||
auth.NextRetryAfter = now.Add(1 * time.Minute)
|
||||
@@ -1462,11 +1485,11 @@ func applyAuthFailureState(auth *Auth, resultErr *Error, retryAfter *time.Durati
|
||||
}
|
||||
|
||||
// nextQuotaCooldown returns the next cooldown duration and updated backoff level for repeated quota errors.
|
||||
func nextQuotaCooldown(prevLevel int) (time.Duration, int) {
|
||||
func nextQuotaCooldown(prevLevel int, disableCooling bool) (time.Duration, int) {
|
||||
if prevLevel < 0 {
|
||||
prevLevel = 0
|
||||
}
|
||||
if quotaCooldownDisabled.Load() {
|
||||
if disableCooling {
|
||||
return 0, prevLevel
|
||||
}
|
||||
cooldown := quotaBackoffBase * time.Duration(1<<prevLevel)
|
||||
@@ -1642,6 +1665,9 @@ func (m *Manager) persist(ctx context.Context, auth *Auth) error {
|
||||
if m.store == nil || auth == nil {
|
||||
return nil
|
||||
}
|
||||
if shouldSkipPersist(ctx) {
|
||||
return nil
|
||||
}
|
||||
if auth.Attributes != nil {
|
||||
if v := strings.ToLower(strings.TrimSpace(auth.Attributes["runtime_only"])); v == "true" {
|
||||
return nil
|
||||
|
||||
97
sdk/cliproxy/auth/conductor_overrides_test.go
Normal file
97
sdk/cliproxy/auth/conductor_overrides_test.go
Normal file
@@ -0,0 +1,97 @@
|
||||
package auth
|
||||
|
||||
import (
|
||||
"context"
|
||||
"testing"
|
||||
"time"
|
||||
)
|
||||
|
||||
func TestManager_ShouldRetryAfterError_RespectsAuthRequestRetryOverride(t *testing.T) {
|
||||
m := NewManager(nil, nil, nil)
|
||||
m.SetRetryConfig(3, 30*time.Second)
|
||||
|
||||
model := "test-model"
|
||||
next := time.Now().Add(5 * time.Second)
|
||||
|
||||
auth := &Auth{
|
||||
ID: "auth-1",
|
||||
Provider: "claude",
|
||||
Metadata: map[string]any{
|
||||
"request_retry": float64(0),
|
||||
},
|
||||
ModelStates: map[string]*ModelState{
|
||||
model: {
|
||||
Unavailable: true,
|
||||
Status: StatusError,
|
||||
NextRetryAfter: next,
|
||||
},
|
||||
},
|
||||
}
|
||||
if _, errRegister := m.Register(context.Background(), auth); errRegister != nil {
|
||||
t.Fatalf("register auth: %v", errRegister)
|
||||
}
|
||||
|
||||
_, maxWait := m.retrySettings()
|
||||
wait, shouldRetry := m.shouldRetryAfterError(&Error{HTTPStatus: 500, Message: "boom"}, 0, []string{"claude"}, model, maxWait)
|
||||
if shouldRetry {
|
||||
t.Fatalf("expected shouldRetry=false for request_retry=0, got true (wait=%v)", wait)
|
||||
}
|
||||
|
||||
auth.Metadata["request_retry"] = float64(1)
|
||||
if _, errUpdate := m.Update(context.Background(), auth); errUpdate != nil {
|
||||
t.Fatalf("update auth: %v", errUpdate)
|
||||
}
|
||||
|
||||
wait, shouldRetry = m.shouldRetryAfterError(&Error{HTTPStatus: 500, Message: "boom"}, 0, []string{"claude"}, model, maxWait)
|
||||
if !shouldRetry {
|
||||
t.Fatalf("expected shouldRetry=true for request_retry=1, got false")
|
||||
}
|
||||
if wait <= 0 {
|
||||
t.Fatalf("expected wait > 0, got %v", wait)
|
||||
}
|
||||
|
||||
_, shouldRetry = m.shouldRetryAfterError(&Error{HTTPStatus: 500, Message: "boom"}, 1, []string{"claude"}, model, maxWait)
|
||||
if shouldRetry {
|
||||
t.Fatalf("expected shouldRetry=false on attempt=1 for request_retry=1, got true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestManager_MarkResult_RespectsAuthDisableCoolingOverride(t *testing.T) {
|
||||
prev := quotaCooldownDisabled.Load()
|
||||
quotaCooldownDisabled.Store(false)
|
||||
t.Cleanup(func() { quotaCooldownDisabled.Store(prev) })
|
||||
|
||||
m := NewManager(nil, nil, nil)
|
||||
|
||||
auth := &Auth{
|
||||
ID: "auth-1",
|
||||
Provider: "claude",
|
||||
Metadata: map[string]any{
|
||||
"disable_cooling": true,
|
||||
},
|
||||
}
|
||||
if _, errRegister := m.Register(context.Background(), auth); errRegister != nil {
|
||||
t.Fatalf("register auth: %v", errRegister)
|
||||
}
|
||||
|
||||
model := "test-model"
|
||||
m.MarkResult(context.Background(), Result{
|
||||
AuthID: "auth-1",
|
||||
Provider: "claude",
|
||||
Model: model,
|
||||
Success: false,
|
||||
Error: &Error{HTTPStatus: 500, Message: "boom"},
|
||||
})
|
||||
|
||||
updated, ok := m.GetByID("auth-1")
|
||||
if !ok || updated == nil {
|
||||
t.Fatalf("expected auth to be present")
|
||||
}
|
||||
state := updated.ModelStates[model]
|
||||
if state == nil {
|
||||
t.Fatalf("expected model state to be present")
|
||||
}
|
||||
if !state.NextRetryAfter.IsZero() {
|
||||
t.Fatalf("expected NextRetryAfter to be zero when disable_cooling=true, got %v", state.NextRetryAfter)
|
||||
}
|
||||
}
|
||||
24
sdk/cliproxy/auth/persist_policy.go
Normal file
24
sdk/cliproxy/auth/persist_policy.go
Normal file
@@ -0,0 +1,24 @@
|
||||
package auth
|
||||
|
||||
import "context"
|
||||
|
||||
type skipPersistContextKey struct{}
|
||||
|
||||
// WithSkipPersist returns a derived context that disables persistence for Manager Update/Register calls.
|
||||
// It is intended for code paths that are reacting to file watcher events, where the file on disk is
|
||||
// already the source of truth and persisting again would create a write-back loop.
|
||||
func WithSkipPersist(ctx context.Context) context.Context {
|
||||
if ctx == nil {
|
||||
ctx = context.Background()
|
||||
}
|
||||
return context.WithValue(ctx, skipPersistContextKey{}, true)
|
||||
}
|
||||
|
||||
func shouldSkipPersist(ctx context.Context) bool {
|
||||
if ctx == nil {
|
||||
return false
|
||||
}
|
||||
v := ctx.Value(skipPersistContextKey{})
|
||||
enabled, ok := v.(bool)
|
||||
return ok && enabled
|
||||
}
|
||||
62
sdk/cliproxy/auth/persist_policy_test.go
Normal file
62
sdk/cliproxy/auth/persist_policy_test.go
Normal file
@@ -0,0 +1,62 @@
|
||||
package auth
|
||||
|
||||
import (
|
||||
"context"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
)
|
||||
|
||||
type countingStore struct {
|
||||
saveCount atomic.Int32
|
||||
}
|
||||
|
||||
func (s *countingStore) List(context.Context) ([]*Auth, error) { return nil, nil }
|
||||
|
||||
func (s *countingStore) Save(context.Context, *Auth) (string, error) {
|
||||
s.saveCount.Add(1)
|
||||
return "", nil
|
||||
}
|
||||
|
||||
func (s *countingStore) Delete(context.Context, string) error { return nil }
|
||||
|
||||
func TestWithSkipPersist_DisablesUpdatePersistence(t *testing.T) {
|
||||
store := &countingStore{}
|
||||
mgr := NewManager(store, nil, nil)
|
||||
auth := &Auth{
|
||||
ID: "auth-1",
|
||||
Provider: "antigravity",
|
||||
Metadata: map[string]any{"type": "antigravity"},
|
||||
}
|
||||
|
||||
if _, err := mgr.Update(context.Background(), auth); err != nil {
|
||||
t.Fatalf("Update returned error: %v", err)
|
||||
}
|
||||
if got := store.saveCount.Load(); got != 1 {
|
||||
t.Fatalf("expected 1 Save call, got %d", got)
|
||||
}
|
||||
|
||||
ctxSkip := WithSkipPersist(context.Background())
|
||||
if _, err := mgr.Update(ctxSkip, auth); err != nil {
|
||||
t.Fatalf("Update(skipPersist) returned error: %v", err)
|
||||
}
|
||||
if got := store.saveCount.Load(); got != 1 {
|
||||
t.Fatalf("expected Save call count to remain 1, got %d", got)
|
||||
}
|
||||
}
|
||||
|
||||
func TestWithSkipPersist_DisablesRegisterPersistence(t *testing.T) {
|
||||
store := &countingStore{}
|
||||
mgr := NewManager(store, nil, nil)
|
||||
auth := &Auth{
|
||||
ID: "auth-1",
|
||||
Provider: "antigravity",
|
||||
Metadata: map[string]any{"type": "antigravity"},
|
||||
}
|
||||
|
||||
if _, err := mgr.Register(WithSkipPersist(context.Background()), auth); err != nil {
|
||||
t.Fatalf("Register(skipPersist) returned error: %v", err)
|
||||
}
|
||||
if got := store.saveCount.Load(); got != 0 {
|
||||
t.Fatalf("expected 0 Save calls, got %d", got)
|
||||
}
|
||||
}
|
||||
@@ -194,6 +194,108 @@ func (a *Auth) ProxyInfo() string {
|
||||
return "via proxy"
|
||||
}
|
||||
|
||||
// DisableCoolingOverride returns the auth-file scoped disable_cooling override when present.
|
||||
// The value is read from metadata key "disable_cooling" (or legacy "disable-cooling").
|
||||
func (a *Auth) DisableCoolingOverride() (bool, bool) {
|
||||
if a == nil || a.Metadata == nil {
|
||||
return false, false
|
||||
}
|
||||
if val, ok := a.Metadata["disable_cooling"]; ok {
|
||||
if parsed, okParse := parseBoolAny(val); okParse {
|
||||
return parsed, true
|
||||
}
|
||||
}
|
||||
if val, ok := a.Metadata["disable-cooling"]; ok {
|
||||
if parsed, okParse := parseBoolAny(val); okParse {
|
||||
return parsed, true
|
||||
}
|
||||
}
|
||||
return false, false
|
||||
}
|
||||
|
||||
// RequestRetryOverride returns the auth-file scoped request_retry override when present.
|
||||
// The value is read from metadata key "request_retry" (or legacy "request-retry").
|
||||
func (a *Auth) RequestRetryOverride() (int, bool) {
|
||||
if a == nil || a.Metadata == nil {
|
||||
return 0, false
|
||||
}
|
||||
if val, ok := a.Metadata["request_retry"]; ok {
|
||||
if parsed, okParse := parseIntAny(val); okParse {
|
||||
if parsed < 0 {
|
||||
parsed = 0
|
||||
}
|
||||
return parsed, true
|
||||
}
|
||||
}
|
||||
if val, ok := a.Metadata["request-retry"]; ok {
|
||||
if parsed, okParse := parseIntAny(val); okParse {
|
||||
if parsed < 0 {
|
||||
parsed = 0
|
||||
}
|
||||
return parsed, true
|
||||
}
|
||||
}
|
||||
return 0, false
|
||||
}
|
||||
|
||||
func parseBoolAny(val any) (bool, bool) {
|
||||
switch typed := val.(type) {
|
||||
case bool:
|
||||
return typed, true
|
||||
case string:
|
||||
trimmed := strings.TrimSpace(typed)
|
||||
if trimmed == "" {
|
||||
return false, false
|
||||
}
|
||||
parsed, err := strconv.ParseBool(trimmed)
|
||||
if err != nil {
|
||||
return false, false
|
||||
}
|
||||
return parsed, true
|
||||
case float64:
|
||||
return typed != 0, true
|
||||
case json.Number:
|
||||
parsed, err := typed.Int64()
|
||||
if err != nil {
|
||||
return false, false
|
||||
}
|
||||
return parsed != 0, true
|
||||
default:
|
||||
return false, false
|
||||
}
|
||||
}
|
||||
|
||||
func parseIntAny(val any) (int, bool) {
|
||||
switch typed := val.(type) {
|
||||
case int:
|
||||
return typed, true
|
||||
case int32:
|
||||
return int(typed), true
|
||||
case int64:
|
||||
return int(typed), true
|
||||
case float64:
|
||||
return int(typed), true
|
||||
case json.Number:
|
||||
parsed, err := typed.Int64()
|
||||
if err != nil {
|
||||
return 0, false
|
||||
}
|
||||
return int(parsed), true
|
||||
case string:
|
||||
trimmed := strings.TrimSpace(typed)
|
||||
if trimmed == "" {
|
||||
return 0, false
|
||||
}
|
||||
parsed, err := strconv.Atoi(trimmed)
|
||||
if err != nil {
|
||||
return 0, false
|
||||
}
|
||||
return parsed, true
|
||||
default:
|
||||
return 0, false
|
||||
}
|
||||
}
|
||||
|
||||
func (a *Auth) AccountInfo() (string, string) {
|
||||
if a == nil {
|
||||
return "", ""
|
||||
|
||||
@@ -124,6 +124,7 @@ func (s *Service) ensureAuthUpdateQueue(ctx context.Context) {
|
||||
}
|
||||
|
||||
func (s *Service) consumeAuthUpdates(ctx context.Context) {
|
||||
ctx = coreauth.WithSkipPersist(ctx)
|
||||
for {
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
@@ -680,6 +681,10 @@ func (s *Service) registerModelsForAuth(a *coreauth.Auth) {
|
||||
if a == nil || a.ID == "" {
|
||||
return
|
||||
}
|
||||
if a.Disabled {
|
||||
GlobalModelRegistry().UnregisterClient(a.ID)
|
||||
return
|
||||
}
|
||||
authKind := strings.ToLower(strings.TrimSpace(a.Attributes["auth_kind"]))
|
||||
if authKind == "" {
|
||||
if kind, _ := a.AccountInfo(); strings.EqualFold(kind, "api_key") {
|
||||
|
||||
@@ -2,6 +2,7 @@ package test
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
@@ -2778,12 +2779,18 @@ func runThinkingTests(t *testing.T, cases []thinkingTestCase) {
|
||||
|
||||
// Verify clear_thinking for iFlow GLM models when enable_thinking=true
|
||||
if tc.to == "iflow" && tc.expectField == "chat_template_kwargs.enable_thinking" && tc.expectValue == "true" {
|
||||
baseModel := thinking.ParseSuffix(tc.model).ModelName
|
||||
isGLM := strings.HasPrefix(strings.ToLower(baseModel), "glm")
|
||||
ctVal := gjson.GetBytes(body, "chat_template_kwargs.clear_thinking")
|
||||
if !ctVal.Exists() {
|
||||
t.Fatalf("expected clear_thinking field not found for GLM model, body=%s", string(body))
|
||||
}
|
||||
if ctVal.Bool() != false {
|
||||
t.Fatalf("clear_thinking: expected false, got %v, body=%s", ctVal.Bool(), string(body))
|
||||
if isGLM {
|
||||
if !ctVal.Exists() {
|
||||
t.Fatalf("expected clear_thinking field not found for GLM model, body=%s", string(body))
|
||||
}
|
||||
if ctVal.Bool() != false {
|
||||
t.Fatalf("clear_thinking: expected false, got %v, body=%s", ctVal.Bool(), string(body))
|
||||
}
|
||||
} else if ctVal.Exists() {
|
||||
t.Fatalf("expected no clear_thinking field for non-GLM enable_thinking model, body=%s", string(body))
|
||||
}
|
||||
}
|
||||
})
|
||||
|
||||
Reference in New Issue
Block a user