Josh LeBlanc and Andrey Nering created
Co-authored-by: Andrey Nering <andreynering@users.noreply.github.com>.github/workflows/update.yml | 1
Taskfile.yaml | 6
cmd/venice/main.go | 272 ++++++++++++
internal/providers/configs/venice.json | 622 ++++++++++++++++++++++++++-
4 files changed, 871 insertions(+), 30 deletions(-)
@@ -24,6 +24,7 @@ jobs:
go run ./cmd/ionet/main.go
go run ./cmd/openrouter/main.go
go run ./cmd/synthetic/main.go
+ go run ./cmd/venice/main.go
go run ./cmd/vercel/main.go
- uses: stefanzweifel/git-auto-commit-action@04702edda442b2e678b25b537cec683a1493fcb9 # v5
with:
@@ -63,6 +63,7 @@ tasks:
- task: gen:ionet
- task: gen:openrouter
- task: gen:synthetic
+ - task: gen:venice
- task: gen:vercel
gen:aihubmix:
@@ -95,6 +96,11 @@ tasks:
cmds:
- go run cmd/synthetic/main.go
+ gen:venice:
+ desc: Generate venice provider configurations
+ cmds:
+ - go run cmd/venice/main.go
+
gen:vercel:
desc: Generate vercel provider configurations
cmds:
@@ -0,0 +1,272 @@
+// Package main provides a command-line tool to fetch models from Venice
+// and generate a configuration file for the provider.
+package main
+
+import (
+ "context"
+ "encoding/json"
+ "fmt"
+ "io"
+ "log"
+ "math"
+ "net/http"
+ "os"
+ "slices"
+ "strings"
+ "time"
+
+ "charm.land/catwalk/pkg/catwalk"
+)
+
+type ModelsResponse struct {
+ Data []VeniceModel `json:"data"`
+}
+
+type VeniceModel struct {
+ Created int64 `json:"created"`
+ ID string `json:"id"`
+ ModelSpec VeniceModelSpec `json:"model_spec"`
+ Object string `json:"object"`
+ OwnedBy string `json:"owned_by"`
+ Type string `json:"type"`
+}
+
+type VeniceModelSpec struct {
+ AvailableContextTokens int64 `json:"availableContextTokens"`
+ Capabilities VeniceModelCapabilities `json:"capabilities"`
+ Constraints VeniceModelConstraints `json:"constraints"`
+ Name string `json:"name"`
+ ModelSource string `json:"modelSource"`
+ Offline bool `json:"offline"`
+ Pricing VeniceModelPricing `json:"pricing"`
+ Traits []string `json:"traits"`
+ Beta bool `json:"beta"`
+}
+
+type VeniceModelCapabilities struct {
+ OptimizedForCode bool `json:"optimizedForCode"`
+ Quantization string `json:"quantization"`
+ SupportsFunctionCalling bool `json:"supportsFunctionCalling"`
+ SupportsReasoning bool `json:"supportsReasoning"`
+ SupportsResponseSchema bool `json:"supportsResponseSchema"`
+ SupportsVision bool `json:"supportsVision"`
+ SupportsWebSearch bool `json:"supportsWebSearch"`
+ SupportsLogProbs bool `json:"supportsLogProbs"`
+}
+
+type VeniceModelConstraints struct {
+ Temperature *VeniceDefaultFloat `json:"temperature"`
+ TopP *VeniceDefaultFloat `json:"top_p"`
+}
+
+type VeniceDefaultFloat struct {
+ Default float64 `json:"default"`
+}
+
+type VeniceModelPricing struct {
+ Input VeniceModelPricingValue `json:"input"`
+ Output VeniceModelPricingValue `json:"output"`
+}
+
+type VeniceModelPricingValue struct {
+ USD float64 `json:"usd"`
+ Diem float64 `json:"diem"`
+}
+
+func fetchVeniceModels(apiEndpoint string) (*ModelsResponse, error) {
+ client := &http.Client{Timeout: 30 * time.Second}
+ url := strings.TrimRight(apiEndpoint, "/") + "/models"
+ req, _ := http.NewRequestWithContext(context.Background(), "GET", url, nil)
+ req.Header.Set("User-Agent", "Crush-Client/1.0")
+
+ if apiKey := strings.TrimSpace(os.Getenv("VENICE_API_KEY")); apiKey != "" && !strings.HasPrefix(apiKey, "$") {
+ req.Header.Set("Authorization", "Bearer "+apiKey)
+ }
+
+ resp, err := client.Do(req)
+ if err != nil {
+ return nil, err //nolint:wrapcheck
+ }
+ defer resp.Body.Close() //nolint:errcheck
+ if resp.StatusCode != 200 {
+ body, _ := io.ReadAll(resp.Body)
+ return nil, fmt.Errorf("status %d: %s", resp.StatusCode, body)
+ }
+
+ var mr ModelsResponse
+ if err := json.NewDecoder(resp.Body).Decode(&mr); err != nil {
+ return nil, err //nolint:wrapcheck
+ }
+ return &mr, nil
+}
+
+func minInt64(a, b int64) int64 {
+ if a < b {
+ return a
+ }
+ return b
+}
+
+func maxInt64(a, b int64) int64 {
+ if a > b {
+ return a
+ }
+ return b
+}
+
+func bestLargeModelID(models []catwalk.Model) string {
+ var best *catwalk.Model
+ for i := range models {
+ m := &models[i]
+
+ if best == nil {
+ best = m
+ continue
+ }
+ mCost := m.CostPer1MIn + m.CostPer1MOut
+ bestCost := best.CostPer1MIn + best.CostPer1MOut
+ if mCost > bestCost {
+ best = m
+ continue
+ }
+ if mCost == bestCost && m.ContextWindow > best.ContextWindow {
+ best = m
+ }
+ }
+ if best == nil {
+ return ""
+ }
+ return best.ID
+}
+
+func bestSmallModelID(models []catwalk.Model) string {
+ var best *catwalk.Model
+ for i := range models {
+ m := &models[i]
+ if best == nil {
+ best = m
+ continue
+ }
+ mCost := m.CostPer1MIn + m.CostPer1MOut
+ bestCost := best.CostPer1MIn + best.CostPer1MOut
+ if mCost < bestCost {
+ best = m
+ continue
+ }
+ if mCost == bestCost && m.ContextWindow < best.ContextWindow {
+ best = m
+ }
+ }
+ if best == nil {
+ return ""
+ }
+ return best.ID
+}
+
+func main() {
+ veniceProvider := catwalk.Provider{
+ Name: "Venice AI",
+ ID: catwalk.InferenceProviderVenice,
+ APIKey: "$VENICE_API_KEY",
+ APIEndpoint: "https://api.venice.ai/api/v1",
+ Type: catwalk.TypeOpenAICompat,
+ Models: []catwalk.Model{},
+ }
+
+ codeOptimizedModels := []catwalk.Model{}
+
+ modelsResp, err := fetchVeniceModels(veniceProvider.APIEndpoint)
+ if err != nil {
+ log.Fatal("Error fetching Venice models:", err)
+ }
+
+ for _, model := range modelsResp.Data {
+ if strings.ToLower(model.Type) != "text" {
+ continue
+ }
+ if model.ModelSpec.Offline {
+ continue
+ }
+ if !model.ModelSpec.Capabilities.SupportsFunctionCalling {
+ continue
+ }
+
+ if model.ModelSpec.Beta {
+ continue
+ }
+
+ contextWindow := model.ModelSpec.AvailableContextTokens
+ if contextWindow <= 0 {
+ continue
+ }
+
+ defaultMaxTokens := minInt64(contextWindow/4, 32768)
+ defaultMaxTokens = maxInt64(defaultMaxTokens, 2048)
+
+ canReason := model.ModelSpec.Capabilities.SupportsReasoning
+ var reasoningLevels []string
+ var defaultReasoning string
+ if canReason {
+ reasoningLevels = []string{"low", "medium", "high"}
+ defaultReasoning = "medium"
+ }
+
+ options := catwalk.ModelOptions{}
+ if model.ModelSpec.Constraints.Temperature != nil {
+ v := model.ModelSpec.Constraints.Temperature.Default
+ if !math.IsNaN(v) {
+ options.Temperature = &v
+ }
+ }
+ if model.ModelSpec.Constraints.TopP != nil {
+ v := model.ModelSpec.Constraints.TopP.Default
+ if !math.IsNaN(v) {
+ options.TopP = &v
+ }
+ }
+
+ m := catwalk.Model{
+ ID: model.ID,
+ Name: model.ModelSpec.Name,
+ CostPer1MIn: model.ModelSpec.Pricing.Input.USD,
+ CostPer1MOut: model.ModelSpec.Pricing.Output.USD,
+ CostPer1MInCached: 0,
+ CostPer1MOutCached: 0,
+ ContextWindow: contextWindow,
+ DefaultMaxTokens: defaultMaxTokens,
+ CanReason: canReason,
+ ReasoningLevels: reasoningLevels,
+ DefaultReasoningEffort: defaultReasoning,
+ SupportsImages: model.ModelSpec.Capabilities.SupportsVision,
+ Options: options,
+ }
+
+ veniceProvider.Models = append(veniceProvider.Models, m)
+ if model.ModelSpec.Capabilities.OptimizedForCode {
+ codeOptimizedModels = append(codeOptimizedModels, m)
+ }
+ }
+
+ candidateModels := veniceProvider.Models
+ if len(codeOptimizedModels) > 0 {
+ candidateModels = codeOptimizedModels
+ }
+
+ veniceProvider.DefaultLargeModelID = bestLargeModelID(candidateModels)
+ veniceProvider.DefaultSmallModelID = bestSmallModelID(candidateModels)
+
+ slices.SortFunc(veniceProvider.Models, func(a catwalk.Model, b catwalk.Model) int {
+ return strings.Compare(a.Name, b.Name)
+ })
+
+ data, err := json.MarshalIndent(veniceProvider, "", " ")
+ if err != nil {
+ log.Fatal("Error marshaling Venice provider:", err)
+ }
+
+ if err := os.WriteFile("internal/providers/configs/venice.json", data, 0o600); err != nil {
+ log.Fatal("Error writing Venice provider config:", err)
+ }
+
+ fmt.Printf("Generated venice.json with %d models\n", len(veniceProvider.Models))
+}
@@ -1,47 +1,405 @@
{
"name": "Venice AI",
"id": "venice",
- "type": "openai-compat",
"api_key": "$VENICE_API_KEY",
"api_endpoint": "https://api.venice.ai/api/v1",
- "default_large_model_id": "qwen3-235b:strip_thinking_response=true",
- "default_small_model_id": "mistral-31-24b",
+ "type": "openai-compat",
+ "default_large_model_id": "claude-opus-4-6",
+ "default_small_model_id": "qwen3-5-35b-a3b",
"models": [
{
- "id": "qwen3-235b:strip_thinking_response=true",
- "name": "Venice Large (qwen3-235b)",
- "cost_per_1m_in": 1.5,
- "cost_per_1m_out": 6,
+ "id": "claude-opus-4-5",
+ "name": "Claude Opus 4.5",
+ "cost_per_1m_in": 6,
+ "cost_per_1m_out": 30,
"cost_per_1m_in_cached": 0,
"cost_per_1m_out_cached": 0,
- "context_window": 131072,
- "default_max_tokens": 50000,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
"can_reason": true,
- "supports_attachments": false
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
},
{
- "id": "qwen3-4b:strip_thinking_response=true",
- "name": "Venice Small (qwen3-4b)",
- "cost_per_1m_in": 0.15,
- "cost_per_1m_out": 0.6,
+ "id": "claude-opus-4-6",
+ "name": "Claude Opus 4.6",
+ "cost_per_1m_in": 6,
+ "cost_per_1m_out": 30,
"cost_per_1m_in_cached": 0,
"cost_per_1m_out_cached": 0,
- "context_window": 32768,
- "default_max_tokens": 25000,
+ "context_window": 1000000,
+ "default_max_tokens": 32768,
"can_reason": true,
- "supports_attachments": false
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
},
{
- "id": "mistral-31-24b",
- "name": "Venice Medium (mistral-31-24b)",
- "cost_per_1m_in": 0.5,
- "cost_per_1m_out": 2,
+ "id": "claude-sonnet-4-5",
+ "name": "Claude Sonnet 4.5",
+ "cost_per_1m_in": 3.75,
+ "cost_per_1m_out": 18.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "claude-sonnet-4-6",
+ "name": "Claude Sonnet 4.6",
+ "cost_per_1m_in": 3.6,
+ "cost_per_1m_out": 18,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 1000000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "zai-org-glm-4.6",
+ "name": "GLM 4.6",
+ "cost_per_1m_in": 0.85,
+ "cost_per_1m_out": 2.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "zai-org-glm-4.7",
+ "name": "GLM 4.7",
+ "cost_per_1m_in": 0.55,
+ "cost_per_1m_out": 2.65,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "zai-org-glm-4.7-flash",
+ "name": "GLM 4.7 Flash",
+ "cost_per_1m_in": 0.125,
+ "cost_per_1m_out": 0.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "olafangensan-glm-4.7-flash-heretic",
+ "name": "GLM 4.7 Flash Heretic",
+ "cost_per_1m_in": 0.14,
+ "cost_per_1m_out": 0.8,
"cost_per_1m_in_cached": 0,
"cost_per_1m_out_cached": 0,
- "context_window": 131072,
- "default_max_tokens": 50000,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "zai-org-glm-5",
+ "name": "GLM 5",
+ "cost_per_1m_in": 1,
+ "cost_per_1m_out": 3.2,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-4o-2024-11-20",
+ "name": "GPT-4o",
+ "cost_per_1m_in": 3.125,
+ "cost_per_1m_out": 12.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
"can_reason": false,
- "supports_attachments": true
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-4o-mini-2024-07-18",
+ "name": "GPT-4o Mini",
+ "cost_per_1m_in": 0.1875,
+ "cost_per_1m_out": 0.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": false,
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-52",
+ "name": "GPT-5.2",
+ "cost_per_1m_in": 2.19,
+ "cost_per_1m_out": 17.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-52-codex",
+ "name": "GPT-5.2 Codex",
+ "cost_per_1m_in": 2.19,
+ "cost_per_1m_out": 17.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-53-codex",
+ "name": "GPT-5.3 Codex",
+ "cost_per_1m_in": 2.19,
+ "cost_per_1m_out": 17.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 400000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "gemini-3-flash-preview",
+ "name": "Gemini 3 Flash Preview",
+ "cost_per_1m_in": 0.7,
+ "cost_per_1m_out": 3.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "gemini-3-pro-preview",
+ "name": "Gemini 3 Pro Preview",
+ "cost_per_1m_in": 2.5,
+ "cost_per_1m_out": 15,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "gemini-3-1-pro-preview",
+ "name": "Gemini 3.1 Pro Preview",
+ "cost_per_1m_in": 2.5,
+ "cost_per_1m_out": 15,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 1000000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "google-gemma-3-27b-it",
+ "name": "Google Gemma 3 27B Instruct",
+ "cost_per_1m_in": 0.12,
+ "cost_per_1m_out": 0.2,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "grok-41-fast",
+ "name": "Grok 4.1 Fast",
+ "cost_per_1m_in": 0.25,
+ "cost_per_1m_out": 0.625,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 1000000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "grok-code-fast-1",
+ "name": "Grok Code Fast 1",
+ "cost_per_1m_in": 0.25,
+ "cost_per_1m_out": 1.87,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "kimi-k2-thinking",
+ "name": "Kimi K2 Thinking",
+ "cost_per_1m_in": 0.75,
+ "cost_per_1m_out": 3.2,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "kimi-k2-5",
+ "name": "Kimi K2.5",
+ "cost_per_1m_in": 0.75,
+ "cost_per_1m_out": 3.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {}
},
{
"id": "llama-3.2-3b",
@@ -50,10 +408,11 @@
"cost_per_1m_out": 0.6,
"cost_per_1m_in_cached": 0,
"cost_per_1m_out_cached": 0,
- "context_window": 131072,
- "default_max_tokens": 25000,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
"can_reason": false,
- "supports_attachments": false
+ "supports_attachments": false,
+ "options": {}
},
{
"id": "llama-3.3-70b",
@@ -62,10 +421,213 @@
"cost_per_1m_out": 2.8,
"cost_per_1m_in_cached": 0,
"cost_per_1m_out_cached": 0,
- "context_window": 65536,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "minimax-m21",
+ "name": "MiniMax M2.1",
+ "cost_per_1m_in": 0.4,
+ "cost_per_1m_out": 1.6,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "minimax-m25",
+ "name": "MiniMax M2.5",
+ "cost_per_1m_in": 0.4,
+ "cost_per_1m_out": 1.6,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 198000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "nvidia-nemotron-3-nano-30b-a3b",
+ "name": "NVIDIA Nemotron 3 Nano 30B",
+ "cost_per_1m_in": 0.075,
+ "cost_per_1m_out": 0.3,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "openai-gpt-oss-120b",
+ "name": "OpenAI GPT OSS 120B",
+ "cost_per_1m_in": 0.07,
+ "cost_per_1m_out": 0.3,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-235b-a22b-instruct-2507",
+ "name": "Qwen 3 235B A22B Instruct 2507",
+ "cost_per_1m_in": 0.15,
+ "cost_per_1m_out": 0.75,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
"default_max_tokens": 32000,
"can_reason": false,
- "supports_attachments": false
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-235b-a22b-thinking-2507",
+ "name": "Qwen 3 235B A22B Thinking 2507",
+ "cost_per_1m_in": 0.45,
+ "cost_per_1m_out": 3.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-coder-480b-a35b-instruct-turbo",
+ "name": "Qwen 3 Coder 480B Turbo",
+ "cost_per_1m_in": 0.35,
+ "cost_per_1m_out": 1.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-coder-480b-a35b-instruct",
+ "name": "Qwen 3 Coder 480b",
+ "cost_per_1m_in": 0.75,
+ "cost_per_1m_out": 3,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-next-80b",
+ "name": "Qwen 3 Next 80b",
+ "cost_per_1m_in": 0.35,
+ "cost_per_1m_out": 1.9,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": false,
+ "options": {}
+ },
+ {
+ "id": "qwen3-5-35b-a3b",
+ "name": "Qwen 3.5 35B A3B",
+ "cost_per_1m_in": 0.3125,
+ "cost_per_1m_out": 1.25,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": true,
+ "options": {
+ "temperature": 1,
+ "top_p": 0.95
+ }
+ },
+ {
+ "id": "qwen3-vl-235b-a22b",
+ "name": "Qwen3 VL 235B",
+ "cost_per_1m_in": 0.25,
+ "cost_per_1m_out": 1.5,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 256000,
+ "default_max_tokens": 32768,
+ "can_reason": false,
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "mistral-31-24b",
+ "name": "Venice Medium",
+ "cost_per_1m_in": 0.5,
+ "cost_per_1m_out": 2,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 128000,
+ "default_max_tokens": 32000,
+ "can_reason": false,
+ "supports_attachments": true,
+ "options": {}
+ },
+ {
+ "id": "qwen3-4b",
+ "name": "Venice Small",
+ "cost_per_1m_in": 0.05,
+ "cost_per_1m_out": 0.15,
+ "cost_per_1m_in_cached": 0,
+ "cost_per_1m_out_cached": 0,
+ "context_window": 32000,
+ "default_max_tokens": 8000,
+ "can_reason": true,
+ "reasoning_levels": [
+ "low",
+ "medium",
+ "high"
+ ],
+ "default_reasoning_effort": "medium",
+ "supports_attachments": false,
+ "options": {}
}
]
-}
+}