You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@maka.apache.org by as...@apache.org on 2026/09/15 08:13:57 UTC
(maka) branch main updated: feat(providers): add Moonshot Global (#5326)
This is an automated email from the ASF dual-hosted git repository.
Astro-Han pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/maka.git
The following commit(s) were added to refs/heads/main by this push:
new 1bdbefa3c5 feat(providers): add Moonshot Global (#5326)
1bdbefa3c5 is described below
commit 1bdbefa3c5dcadda73abcb9eac7f70b3cb7e21c8
Author: AstroHan <le...@outlook.com>
AuthorDate: Tue Sep 15 16:13:52 2026 +0800
feat(providers): add Moonshot Global (#5326)
Add Moonshot Global through the existing OpenAI Responses adapter, using
the models.dev moonshotai provider for its endpoint, models, and pricing.
Keep the existing Moonshot China provider and saved connections unchanged.
Refresh the models.dev snapshot and use the shared Kimi logo for mainland,
international, and Coding Plan entries. Cover the international Responses
tool loop and remove catalog-dependent assertions that block valid refreshes.
Generated-by: Codex
---
.../src/renderer/settings/provider-brand-marks.tsx | 8 +-
.../src/renderer/settings/provider-display-copy.ts | 5 +
packages/core/src/__tests__/model-metadata.test.ts | 4 -
.../__tests__/provider-catalog-contract.test.ts | 13 +
packages/core/src/models-dev-projection.ts | 1 +
packages/core/src/provider-registry.ts | 20 +
packages/runtime/src/__tests__/cost.test.ts | 6 +
.../src/__tests__/model-factory-thinking.test.ts | 1 -
.../src/__tests__/provider-contract-overrides.ts | 18 +
.../model-metadata/models-dev-api.snapshot.json | 3445 ++++++++++++++------
10 files changed, 2597 insertions(+), 924 deletions(-)
diff --git a/apps/desktop/src/renderer/settings/provider-brand-marks.tsx b/apps/desktop/src/renderer/settings/provider-brand-marks.tsx
index 4e9d095e58..67e1aa18e8 100644
--- a/apps/desktop/src/renderer/settings/provider-brand-marks.tsx
+++ b/apps/desktop/src/renderer/settings/provider-brand-marks.tsx
@@ -317,10 +317,11 @@ function DeepSeek(): ReactElement {
);
}
-function Moonshot(): ReactElement {
+function Kimi(): ReactElement {
return (
<svg viewBox="0 0 24 24" role="img" fill="currentColor" fillRule="evenodd" xmlns="http://www.w3.org/2000/svg">
- <path d="M1.052 16.916l9.539 2.552a21.007 21.007 0 00.06 2.033l5.956 1.593a11.997 11.997 0 01-5.586.865l-.18-.016-.044-.004-.084-.009-.094-.01a11.605 11.605 0 01-.157-.02l-.107-.014-.11-.016a11.962 11.962 0 01-.32-.051l-.042-.008-.075-.013-.107-.02-.07-.015-.093-.019-.075-.016-.095-.02-.097-.023-.094-.022-.068-.017-.088-.022-.09-.024-.095-.025-.082-.023-.109-.03-.062-.02-.084-.025-.093-.028-.105-.034-.058-.019-.08-.026-.09-.031-.066-.024-.044-.015-.068-.025-.101-.037-.057-.022-.08- [...]
+ <path d="M21.846 0a1.923 1.923 0 110 3.846H20.15a.226.226 0 01-.227-.226V1.923C19.923.861 20.784 0 21.846 0z" />
+ <path d="M11.065 11.199l7.257-7.2c.137-.136.06-.41-.116-.41H14.3a.164.164 0 00-.117.051l-7.82 7.756c-.122.12-.302.013-.302-.179V3.82c0-.127-.083-.23-.185-.23H3.186c-.103 0-.186.103-.186.23V19.77c0 .128.083.23.186.23h2.69c.103 0 .186-.102.186-.23v-3.25c0-.069.025-.135.069-.178l2.424-2.406a.158.158 0 01.205-.023l6.484 4.772a7.677 7.677 0 003.453 1.283c.108.012.2-.095.2-.23v-3.06c0-.117-.07-.212-.164-.227a5.028 5.028 0 01-2.027-.807l-5.613-4.064c-.117-.078-.132-.279-.028-.381z" />
</svg>
);
}
@@ -412,8 +413,9 @@ export function ProviderBrandMark({ type }: { type: ProviderType }): ReactElemen
case 'deepseek':
return <DeepSeek />;
case 'moonshot':
+ case 'moonshot-global':
case 'kimi-coding-plan':
- return <Moonshot />;
+ return <Kimi />;
case 'xiaomi-token-plan-cn':
case 'xiaomi-token-plan-sgp':
case 'xiaomi-token-plan-ams':
diff --git a/apps/desktop/src/renderer/settings/provider-display-copy.ts b/apps/desktop/src/renderer/settings/provider-display-copy.ts
index 744a62fdbb..6730ffdee1 100644
--- a/apps/desktop/src/renderer/settings/provider-display-copy.ts
+++ b/apps/desktop/src/renderer/settings/provider-display-copy.ts
@@ -74,6 +74,11 @@ export const PROVIDER_DISPLAY_COPY = {
'zh-TW': { name: 'Moonshot', description: 'Moonshot 官方 API 連線', badge: 'API' },
en: { name: 'Moonshot', description: 'Official Moonshot API access.', badge: 'API' },
},
+ 'moonshot-global': {
+ 'zh-CN': { name: 'Moonshot 国际版', description: 'Moonshot 国际版官方 API 接入', badge: 'API' },
+ 'zh-TW': { name: 'Moonshot 國際版', description: 'Moonshot 國際版官方 API 連線', badge: 'API' },
+ en: { name: 'Moonshot Global', description: 'Official Moonshot international API access.', badge: 'API' },
+ },
'zai-coding-plan': {
'zh-CN': { name: 'Z.AI Coding Plan', description: '智谱 · OpenAI 兼容', badge: 'Coding' },
'zh-TW': { name: 'Z.AI Coding Plan', description: '智譜 · OpenAI 相容', badge: 'Coding' },
diff --git a/packages/core/src/__tests__/model-metadata.test.ts b/packages/core/src/__tests__/model-metadata.test.ts
index 8609efcfb8..90bfbc03ab 100644
--- a/packages/core/src/__tests__/model-metadata.test.ts
+++ b/packages/core/src/__tests__/model-metadata.test.ts
@@ -167,10 +167,6 @@ describe('deepseek v4 flash vision exp metadata regression', () => {
assert.equal(metadata.displayName, 'DeepSeek-V4-Flash-Vision-Exp');
assert.equal(metadata.capabilities?.vision, true);
assert.equal(resolveModelVisionSupport('deepseek', discovered, modelId), true);
- assert.equal(
- resolveModelVisionSupport('deepseek', [{ id: 'deepseek-v4-flash' }], 'deepseek-v4-flash'),
- false,
- );
});
});
diff --git a/packages/core/src/__tests__/provider-catalog-contract.test.ts b/packages/core/src/__tests__/provider-catalog-contract.test.ts
index 91b02c10c8..16717ffe40 100644
--- a/packages/core/src/__tests__/provider-catalog-contract.test.ts
+++ b/packages/core/src/__tests__/provider-catalog-contract.test.ts
@@ -48,6 +48,19 @@ describe('provider connection slug derivation contract', () => {
});
});
+describe('Moonshot provider regions', () => {
+ it('offers the international API as its own provider', () => {
+ const global = PROVIDER_REGISTRY['moonshot-global'];
+ assert.equal(global.baseUrl, 'https://api.moonshot.ai/v1');
+ assert.equal(global.category, 'overseas');
+ assert.deepEqual(global.runtimeAdapter, {
+ kind: 'openai',
+ apiProtocol: 'openai-responses',
+ });
+ assert.ok(global.fallbackModels.includes('kimi-k3'));
+ });
+});
+
describe('provider catalog contract — structural invariants over CATALOG_PROVIDER_TYPES', () => {
it('exposes an endpoint source that passes the production baseUrl gate', () => {
for (const type of CATALOG_PROVIDER_TYPES) {
diff --git a/packages/core/src/models-dev-projection.ts b/packages/core/src/models-dev-projection.ts
index bf31492690..e4d4d09ea7 100644
--- a/packages/core/src/models-dev-projection.ts
+++ b/packages/core/src/models-dev-projection.ts
@@ -61,6 +61,7 @@ export const MODELS_DEV_PROVIDERS = {
'minimax-coding-plan': 'minimax-coding-plan',
mistral: 'mistral',
moonshot: 'moonshotai-cn',
+ 'moonshot-global': 'moonshotai',
nvidia: 'nvidia',
'ollama-cloud': 'ollama-cloud',
openai: 'openai',
diff --git a/packages/core/src/provider-registry.ts b/packages/core/src/provider-registry.ts
index ebf0f74356..1fc30e4b47 100644
--- a/packages/core/src/provider-registry.ts
+++ b/packages/core/src/provider-registry.ts
@@ -647,6 +647,13 @@ const moonshotModelIds = toolCallingModelIds('Moonshot', GENERATED_MODELS_DEV_ME
'kimi-k2.6',
'kimi-k2.7-code',
]).filter((id) => GENERATED_MODELS_DEV_METADATA.moonshot[id]?.lifecycle !== 'deprecated');
+const moonshotGlobal = GENERATED_MODELS_DEV_PROVIDER_FACTS['moonshot-global'];
+if (!moonshotGlobal.api) throw new Error('models.dev Moonshot Global provider is missing its API');
+const moonshotGlobalModelIds = toolCallingModelIds(
+ 'Moonshot Global',
+ GENERATED_MODELS_DEV_METADATA['moonshot-global'],
+ ['kimi-k3'],
+).filter((id) => GENERATED_MODELS_DEV_METADATA['moonshot-global'][id]?.lifecycle !== 'deprecated');
const cloudflareWorkersAi = GENERATED_MODELS_DEV_PROVIDER_FACTS['cloudflare-workers-ai'];
if (cloudflareWorkersAi.id !== 'cloudflare-workers-ai') {
throw new Error(
@@ -945,6 +952,19 @@ const providerRegistry = {
signupUrl: 'https://platform.kimi.com/console/api-keys',
catalogOrder: 4,
},
+ 'moonshot-global': {
+ label: 'Moonshot Global',
+ baseUrl: moonshotGlobal.api,
+ authKind: 'api_key',
+ fallbackModels: moonshotGlobalModelIds,
+ status: 'ready',
+ runtimeAdapter: { kind: 'openai', apiProtocol: 'openai-responses' },
+ modelDiscovery: { kind: 'protocol' },
+ category: 'overseas',
+ catalogGroup: 'api',
+ signupUrl: 'https://platform.kimi.ai/console/api-keys',
+ catalogOrder: 4.1,
+ },
'zai-coding-plan': {
label: 'Z.AI Coding Plan',
menuLabel: 'Z.AI',
diff --git a/packages/runtime/src/__tests__/cost.test.ts b/packages/runtime/src/__tests__/cost.test.ts
index 698340ac41..4688fdf588 100644
--- a/packages/runtime/src/__tests__/cost.test.ts
+++ b/packages/runtime/src/__tests__/cost.test.ts
@@ -20,11 +20,17 @@
import assert from 'node:assert/strict';
import { describe, test } from 'node:test';
+import { getBuiltinPricing } from '../telemetry/builtin-pricing.js';
import { computeCost } from '../telemetry/cost.js';
import { recordLlmCall } from '../telemetry/record-llm-call.js';
import { recordToolInvocation } from '../telemetry/record-tool-invocation.js';
import type { PersistedLlmCallRecord, PersistedToolInvocationRecord } from '../telemetry/types.js';
+test('Moonshot Global pricing comes from the models.dev snapshot', () => {
+ const pricing = getBuiltinPricing('moonshot-global:kimi-k3');
+ assert.ok(pricing && pricing.inputUsdPer1M > 0 && pricing.outputUsdPer1M > 0);
+});
+
describe('computeCost', () => {
test('charges full input price only for cache-miss input', () => {
const cost = computeCost(
diff --git a/packages/runtime/src/__tests__/model-factory-thinking.test.ts b/packages/runtime/src/__tests__/model-factory-thinking.test.ts
index 7d60cb5262..ce5752f024 100644
--- a/packages/runtime/src/__tests__/model-factory-thinking.test.ts
+++ b/packages/runtime/src/__tests__/model-factory-thinking.test.ts
@@ -503,7 +503,6 @@ describe('buildProviderOptions: thinking level', () => {
}
}
- assert.equal(activeClaudeModels.length, 16);
assert.ok(
activeClaudeModels.some(
({ connection, modelId }) =>
diff --git a/packages/runtime/src/__tests__/provider-contract-overrides.ts b/packages/runtime/src/__tests__/provider-contract-overrides.ts
index a3b1af6310..13891c5d69 100644
--- a/packages/runtime/src/__tests__/provider-contract-overrides.ts
+++ b/packages/runtime/src/__tests__/provider-contract-overrides.ts
@@ -169,6 +169,24 @@ export const PROVIDER_CONTRACT_OVERRIDE_BINDINGS: readonly ProviderContractOverr
statelessReasoning: true,
}),
},
+ {
+ keys: [
+ 'moonshot-global:exact-model-id',
+ 'moonshot-global:tool-loop',
+ 'moonshot-global:reasoning-replay',
+ ],
+ title: 'Moonshot Global preserves Kimi model ids and reasoning across a Responses tool loop',
+ run: () =>
+ runOpenAIResponsesWire({
+ providerType: 'moonshot-global',
+ slug: 'moonshot-global',
+ name: 'Moonshot Global',
+ basePath: '/v1',
+ modelId: 'kimi-k3',
+ apiKey: 'moonshot-global-test-key',
+ statelessReasoning: true,
+ }),
+ },
{
keys: ['alibaba-token-plan-cn:reasoning-replay', 'alibaba-token-plan:reasoning-replay'],
title: 'Alibaba Token Plan replays plaintext summary items on its Responses wire',
diff --git a/scripts/model-metadata/models-dev-api.snapshot.json b/scripts/model-metadata/models-dev-api.snapshot.json
index c7c5161db3..ece05db73e 100644
--- a/scripts/model-metadata/models-dev-api.snapshot.json
+++ b/scripts/model-metadata/models-dev-api.snapshot.json
@@ -3,11 +3,11 @@
"sourceUrl": "https://models.dev/api.json",
"origin": {
"kind": "models-dev-response",
- "retrievedAt": "2026-09-01T10:29:38.000Z",
- "etag": "W/\"7559b5b3701539769901f1e20e89315d\"",
- "responseSha256": "7559b5b3701539769901f1e20e89315d94975a7b289646590d6f854a21e12f17"
+ "retrievedAt": "2026-09-15T03:05:08.000Z",
+ "etag": "W/\"153362e16f62e2c0b4bbb4d66949e9c5\"",
+ "responseSha256": "153362e16f62e2c0b4bbb4d66949e9c5ccc050e0dd7dbd325bbf434f2a8fcbf1"
},
- "projectionSha256": "85b25f334b65929079ea6dab9b0a2a66f433e3d66710fcd1aadff5f611b37345",
+ "projectionSha256": "b53ff13413fc4547fa0baa488edc03339e7b78ebe5fb6898d8c9dd4d00d059c3",
"projection": {
"metadata": {
"anthropic": {
@@ -32,6 +32,28 @@
"output": ["text"]
}
},
+ "claude-fable-5-1": {
+ "displayName": "Claude Fable 5.1",
+ "description": "Claude model for demanding reasoning and long-horizon agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-06",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-01",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"claude-haiku-4-5": {
"displayName": "Claude Haiku 4.5 (latest)",
"description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat",
@@ -3627,6 +3649,30 @@
"output": ["text"]
}
},
+ "deepseek-v4.1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"glm-5": {
"displayName": "GLM-5",
"description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows",
@@ -4010,7 +4056,7 @@
"qwen3.8-max-preview": {
"displayName": "Qwen3.8 Max Preview",
"description": "Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows",
- "lifecycle": "beta",
+ "lifecycle": "deprecated",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
"structuredOutput": true,
@@ -4568,7 +4614,7 @@
"qwen3.8-max-preview": {
"displayName": "Qwen3.8 Max Preview",
"description": "Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows",
- "lifecycle": "beta",
+ "lifecycle": "deprecated",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
"structuredOutput": true,
@@ -4625,45 +4671,45 @@
}
},
"cerebras": {
- "gemma-4-31b": {
- "displayName": "Gemma 4 31B IT",
- "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
- "lifecycle": "beta",
+ "gpt-oss-120b": {
+ "displayName": "GPT OSS 120B",
+ "description": "Open GPT reasoning model for self-hosted agents and controllable deployments",
+ "lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 40960,
"structuredOutput": true,
- "lastUpdated": "2026-07-01",
+ "lastUpdated": "2026-06-10",
"capabilities": {
- "vision": true,
+ "vision": false,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["none", "low", "medium", "high"]
+ "efforts": ["low", "medium", "high"]
},
"modalities": {
- "input": ["text", "image"],
+ "input": ["text"],
"output": ["text"]
}
},
- "gpt-oss-120b": {
- "displayName": "GPT OSS 120B",
- "description": "Open GPT reasoning model for self-hosted agents and controllable deployments",
+ "qwen-3.8-27b": {
+ "displayName": "Qwen3.8 27B",
+ "description": "Dense 27B vision-language model for coding, agent tasks, and image and video understanding",
"lifecycle": "active",
- "contextWindow": 131072,
- "maxOutputTokens": 40960,
+ "contextWindow": 65536,
+ "maxOutputTokens": 32768,
"structuredOutput": true,
- "lastUpdated": "2026-06-10",
+ "lastUpdated": "2026-09-03",
"capabilities": {
- "vision": false,
+ "vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["low", "medium", "high"]
+ "efforts": ["none", "low", "medium", "high"]
},
"modalities": {
- "input": ["text"],
+ "input": ["text", "image"],
"output": ["text"]
}
}
@@ -5447,8 +5493,7 @@
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["low", "medium", "high"],
- "toggle": true
+ "efforts": ["low", "medium", "high"]
},
"modalities": {
"input": ["text"],
@@ -5675,6 +5720,27 @@
"output": ["text"]
}
},
+ "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {
+ "displayName": "DeepSeek V4 Flash Vision Exp",
+ "description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-21",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"deepseek-ai/DeepSeek-V4-Pro": {
"displayName": "DeepSeek V4 Pro",
"description": "Open MoE flagship with million-token context for coding and long agent runs",
@@ -5720,6 +5786,85 @@
"output": ["text"]
}
},
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-12b-it": {
+ "displayName": "Gemma 3 12B IT",
+ "description": "Open multimodal Gemma instruction model for multilingual text generation and image understanding",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-27b-it": {
+ "displayName": "Gemma 3 27B IT",
+ "description": "Largest open Gemma 3 instruction model for multilingual text generation and visual understanding",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-4b-it": {
+ "displayName": "Gemma 3 4B IT",
+ "description": "Open multimodal Gemma instruction model for efficient text generation and image understanding",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"google/gemma-4-26B-A4B-it": {
"displayName": "Gemma 4 26B A4B IT",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
@@ -5858,7 +6003,7 @@
"MiniMaxAI/MiniMax-M2.7": {
"displayName": "MiniMax-M2.7",
"description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 196608,
"maxOutputTokens": 131072,
"lastUpdated": "2026-03-18",
@@ -5893,7 +6038,7 @@
"moonshotai/Kimi-K2.5": {
"displayName": "Kimi K2.5",
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 262144,
"maxOutputTokens": 32768,
"knowledgeCutoff": "2025-01",
@@ -6391,7 +6536,25 @@
"toggle": true
},
"modalities": {
- "input": ["text", "image"],
+ "input": ["text", "image", "video"],
+ "output": ["text"]
+ }
+ },
+ "Qwen/Qwen3.8-Flash": {
+ "displayName": "Qwen3.8 Flash",
+ "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-26",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image", "video"],
"output": ["text"]
}
},
@@ -6436,7 +6599,8 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 64000,
+ "inputLimit": 192000,
+ "maxOutputTokens": 128000,
"structuredOutput": true,
"lastUpdated": "2026-07-06",
"capabilities": {
@@ -6575,7 +6739,7 @@
"zai-org/GLM-4.7-Flash": {
"displayName": "GLM-4.7-Flash",
"description": "Efficient GLM model for fast reasoning, coding, and agent workflows",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 202752,
"maxOutputTokens": 16384,
"knowledgeCutoff": "2025-04",
@@ -6594,7 +6758,7 @@
"zai-org/GLM-5": {
"displayName": "GLM-5",
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 202752,
"maxOutputTokens": 16384,
"knowledgeCutoff": "2025-12",
@@ -6701,17 +6865,40 @@
}
},
"deepseek": {
+ "deepseek-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"deepseek-v4-flash": {
"displayName": "DeepSeek V4 Flash",
- "description": "Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
"lifecycle": "active",
"contextWindow": 1000000,
"maxOutputTokens": 384000,
"knowledgeCutoff": "2025-05",
"structuredOutput": true,
- "lastUpdated": "2026-07-31",
+ "lastUpdated": "2026-09-10",
"capabilities": {
- "vision": false,
+ "vision": true,
"reasoning": true,
"functionCalling": true
},
@@ -6720,18 +6907,19 @@
"toggle": true
},
"modalities": {
- "input": ["text"],
+ "input": ["text", "image"],
"output": ["text"]
}
},
"deepseek-v4-flash-vision-exp": {
"displayName": "DeepSeek V4 Flash Vision Exp",
- "description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
- "lifecycle": "beta",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
"contextWindow": 1000000,
"maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
"structuredOutput": true,
- "lastUpdated": "2026-08-21",
+ "lastUpdated": "2026-09-10",
"capabilities": {
"vision": true,
"reasoning": true,
@@ -6793,6 +6981,28 @@
"output": ["text"]
}
},
+ "accounts/fireworks/models/deepseek-v4-flash-vision-exp": {
+ "displayName": "DeepSeek V4 Flash Vision Exp",
+ "description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-21",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"accounts/fireworks/models/deepseek-v4-pro-0813": {
"displayName": "DeepSeek V4 Pro 0813",
"description": "DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes",
@@ -6815,6 +7025,29 @@
"output": ["text"]
}
},
+ "accounts/fireworks/models/deepseek-v4p1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"accounts/fireworks/models/glm-5p2": {
"displayName": "GLM 5.2",
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
@@ -6840,17 +7073,17 @@
"displayName": "GLM 5.3",
"description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
"lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 131072,
+ "contextWindow": 1048573,
+ "maxOutputTokens": 262144,
"structuredOutput": true,
- "lastUpdated": "2026-08-28",
+ "lastUpdated": "2026-09-07",
"capabilities": {
"vision": false,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["high", "max"]
+ "efforts": ["low", "high", "max"]
},
"modalities": {
"input": ["text"],
@@ -6861,17 +7094,17 @@
"displayName": "GLM 5.3 Flash",
"description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
"lifecycle": "active",
- "contextWindow": 1000000,
+ "contextWindow": 1048573,
"maxOutputTokens": 131072,
"structuredOutput": true,
- "lastUpdated": "2026-08-26",
+ "lastUpdated": "2026-09-07",
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["high", "max"]
+ "efforts": ["low", "high", "max"]
},
"modalities": {
"input": ["text", "image", "video", "pdf"],
@@ -6997,6 +7230,24 @@
"output": ["text"]
}
},
+ "accounts/fireworks/models/mistral-large-3-fp8": {
+ "displayName": "Mistral Large 3 675B Instruct 2512",
+ "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 262144,
+ "knowledgeCutoff": "2024-11",
+ "lastUpdated": "2025-12-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"accounts/fireworks/models/muse-glimmer-30b": {
"displayName": "Muse Glimmer 30B",
"description": "Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.",
@@ -7081,6 +7332,27 @@
"output": ["text"]
}
},
+ "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
+ "displayName": "Qwen3.8 2.4T A95B",
+ "description": "Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-12",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "xhigh"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"accounts/fireworks/models/qwen3p8-max": {
"displayName": "Qwen3.8 Max",
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
@@ -7122,6 +7394,27 @@
"output": ["text"]
}
},
+ "accounts/fireworks/routers/glm-5p3-fast": {
+ "displayName": "GLM 5.3 Fast",
+ "description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
+ "lifecycle": "active",
+ "contextWindow": 1048572,
+ "maxOutputTokens": 262144,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-07",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"accounts/fireworks/routers/kimi-k3-fast": {
"displayName": "Kimi K3 Fast",
"description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
@@ -7167,61 +7460,41 @@
"output": ["text"]
}
},
- "claude-haiku-4.5": {
- "displayName": "Claude Haiku 4.5 (latest)",
- "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat",
+ "claude-fable-5.1": {
+ "displayName": "Claude Fable 5.1",
+ "description": "Claude model for demanding reasoning and long-horizon agentic work",
"lifecycle": "active",
- "contextWindow": 200000,
- "inputLimit": 136000,
- "maxOutputTokens": 64000,
- "knowledgeCutoff": "2025-02-28",
- "lastUpdated": "2025-10-15",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-06",
+ "lastUpdated": "2026-09-01",
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
- "claude-opus-4.5": {
- "displayName": "Claude Opus 4.5 (latest)",
- "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
- "lifecycle": "active",
- "contextWindow": 200000,
- "inputLimit": 168000,
- "maxOutputTokens": 32000,
- "knowledgeCutoff": "2025-05",
- "lastUpdated": "2025-11-24",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
},
"modalities": {
"input": ["text", "image", "pdf"],
"output": ["text"]
}
},
- "claude-opus-4.6": {
- "displayName": "Claude Opus 4.6",
- "description": "High-end Claude for difficult coding, planning, and slower expert reasoning",
+ "claude-haiku-4.5": {
+ "displayName": "Claude Haiku 4.5 (latest)",
+ "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat",
"lifecycle": "active",
"contextWindow": 200000,
- "inputLimit": 168000,
- "maxOutputTokens": 32000,
- "knowledgeCutoff": "2025-05-31",
- "lastUpdated": "2026-03-13",
+ "inputLimit": 136000,
+ "maxOutputTokens": 64000,
+ "knowledgeCutoff": "2025-02-28",
+ "lastUpdated": "2025-10-15",
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
- "thinkingOptions": {
- "efforts": ["low", "medium", "high", "max"]
- },
"modalities": {
"input": ["text", "image", "pdf"],
"output": ["text"]
@@ -7294,44 +7567,6 @@
"output": ["text"]
}
},
- "claude-sonnet-4": {
- "displayName": "Claude Sonnet 4 (latest)",
- "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
- "lifecycle": "active",
- "contextWindow": 216000,
- "inputLimit": 128000,
- "maxOutputTokens": 16000,
- "knowledgeCutoff": "2025-03-31",
- "lastUpdated": "2025-05-22",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
- "claude-sonnet-4.5": {
- "displayName": "Claude Sonnet 4.5 (latest)",
- "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
- "lifecycle": "active",
- "contextWindow": 200000,
- "inputLimit": 168000,
- "maxOutputTokens": 32000,
- "knowledgeCutoff": "2025-07-31",
- "lastUpdated": "2025-09-29",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
"claude-sonnet-4.6": {
"displayName": "Claude Sonnet 4.6",
"description": "Claude workhorse for coding agents, careful analysis, and production cost control",
@@ -7375,29 +7610,6 @@
"output": ["text"]
}
},
- "gemini-3.1-pro-preview": {
- "displayName": "Gemini 3.1 Pro Preview",
- "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving",
- "lifecycle": "active",
- "contextWindow": 1000000,
- "inputLimit": 936000,
- "maxOutputTokens": 64000,
- "knowledgeCutoff": "2025-01",
- "structuredOutput": true,
- "lastUpdated": "2026-02-19",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "efforts": ["low", "medium", "high"]
- },
- "modalities": {
- "input": ["text", "image", "video", "audio", "pdf"],
- "output": ["text"]
- }
- },
"gemini-3.5-flash": {
"displayName": "Gemini 3.5 Flash",
"description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
@@ -7467,23 +7679,25 @@
"output": ["text"]
}
},
- "gpt-4.1": {
- "displayName": "GPT-4.1",
- "description": "Long-lived GPT workhorse for coding, instruction following, and production apps",
+ "gemini-3.8-flash": {
+ "displayName": "Gemini 3.8 Flash",
+ "description": "Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows",
"lifecycle": "active",
- "contextWindow": 128000,
- "inputLimit": 128000,
- "maxOutputTokens": 16384,
- "knowledgeCutoff": "2024-04",
+ "contextWindow": 1000000,
+ "inputLimit": 936000,
+ "maxOutputTokens": 64000,
"structuredOutput": true,
- "lastUpdated": "2025-04-14",
+ "lastUpdated": "2026-09-02",
"capabilities": {
"vision": true,
- "reasoning": false,
+ "reasoning": true,
"functionCalling": true
},
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high"]
+ },
"modalities": {
- "input": ["text", "image", "pdf"],
+ "input": ["text", "image"],
"output": ["text"]
}
},
@@ -7510,46 +7724,6 @@
"output": ["text"]
}
},
- "gpt-5.2": {
- "displayName": "GPT-5.2",
- "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work",
- "lifecycle": "active",
- "contextWindow": 400000,
- "inputLimit": 272000,
- "maxOutputTokens": 128000,
- "knowledgeCutoff": "2025-08-31",
- "structuredOutput": true,
- "lastUpdated": "2025-12-11",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text", "image"],
- "output": ["text"]
- }
- },
- "gpt-5.2-codex": {
- "displayName": "GPT-5.2 Codex",
- "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents",
- "lifecycle": "active",
- "contextWindow": 400000,
- "inputLimit": 272000,
- "maxOutputTokens": 128000,
- "knowledgeCutoff": "2025-08-31",
- "structuredOutput": true,
- "lastUpdated": "2025-12-11",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
"gpt-5.3-codex": {
"displayName": "GPT-5.3 Codex",
"description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work",
@@ -7731,6 +7905,29 @@
"output": ["text"]
}
},
+ "gpt-6-astra": {
+ "displayName": "GPT-6 Astra",
+ "description": "GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-04-30",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"grok-4.5": {
"displayName": "Grok 4.5",
"description": "xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk",
@@ -8394,6 +8591,27 @@
"output": ["text"]
}
},
+ "gemini-3.8-flash": {
+ "displayName": "Gemini 3.8 Flash",
+ "description": "Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "audio", "pdf"],
+ "output": ["text"]
+ }
+ },
"gemini-embedding-001": {
"displayName": "Gemini Embedding 001",
"description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
@@ -9071,6 +9289,27 @@
"output": ["text"]
}
},
+ "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {
+ "displayName": "DeepSeek V4 Flash Vision Exp",
+ "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-21",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"deepseek-ai/DeepSeek-V4-Pro": {
"displayName": "DeepSeek V4 Pro",
"description": "Open MoE flagship with million-token context for coding and long agent runs",
@@ -9114,6 +9353,85 @@
"output": ["text"]
}
},
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-12b-it": {
+ "displayName": "Gemma 3 12B IT",
+ "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-27b-it": {
+ "displayName": "Gemma 3 27B IT",
+ "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "google/gemma-3-4b-it": {
+ "displayName": "Gemma 3 4B IT",
+ "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 131072,
+ "knowledgeCutoff": "2024-08",
+ "structuredOutput": true,
+ "lastUpdated": "2025-03-12",
+ "capabilities": {
+ "vision": true,
+ "reasoning": false,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"google/gemma-4-26B-A4B-it": {
"displayName": "Gemma 4 26B A4B IT",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
@@ -9941,7 +10259,8 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 64000,
+ "inputLimit": 192000,
+ "maxOutputTokens": 128000,
"structuredOutput": true,
"lastUpdated": "2026-07-06",
"capabilities": {
@@ -10330,20 +10649,22 @@
}
},
"kimi-for-coding": {
- "displayName": "Kimi K2.7 Code",
- "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
+ "displayName": "kimi-for-coding",
+ "description": "Kimi coding model with more efficient thinking and up to 1M context, available through Kimi Code",
"lifecycle": "active",
- "contextWindow": 262144,
+ "contextWindow": 1048576,
"maxOutputTokens": 32768,
- "knowledgeCutoff": "2025-01",
- "structuredOutput": true,
- "lastUpdated": "2026-06-12",
+ "lastUpdated": "2026-09-11",
"isFree": true,
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
"modalities": {
"input": ["text", "image", "video"],
"output": ["text"]
@@ -11376,118 +11697,90 @@
}
},
"moonshot": {
- "kimi-k2-0711-preview": {
- "displayName": "Kimi K2 0711",
- "description": "Kimi model for long-context chat, coding, and agentic reasoning",
- "lifecycle": "active",
- "contextWindow": 131072,
- "maxOutputTokens": 16384,
- "knowledgeCutoff": "2024-10",
- "lastUpdated": "2025-07-14",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
- "kimi-k2-0905-preview": {
- "displayName": "Kimi K2 0905",
- "description": "Kimi model for long-context chat, coding, and agentic reasoning",
- "lifecycle": "active",
- "contextWindow": 262144,
- "maxOutputTokens": 262144,
- "knowledgeCutoff": "2024-10",
- "lastUpdated": "2025-09-05",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
- "kimi-k2-thinking": {
- "displayName": "Kimi K2 Thinking",
- "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions",
+ "kimi-k2.6": {
+ "displayName": "Kimi K2.6",
+ "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context",
"lifecycle": "active",
"contextWindow": 262144,
"maxOutputTokens": 262144,
- "knowledgeCutoff": "2024-08",
- "lastUpdated": "2025-11-06",
+ "knowledgeCutoff": "2025-01",
+ "structuredOutput": true,
+ "lastUpdated": "2026-04-21",
"capabilities": {
- "vision": false,
+ "vision": true,
"reasoning": true,
"functionCalling": true
},
+ "thinkingOptions": {
+ "toggle": true
+ },
"modalities": {
- "input": ["text"],
+ "input": ["text", "image", "video"],
"output": ["text"]
}
},
- "kimi-k2-thinking-turbo": {
- "displayName": "Kimi K2 Thinking Turbo",
- "description": "Kimi reasoning model for long-horizon research, planning, and tool use",
+ "kimi-k2.7-code": {
+ "displayName": "Kimi K2.7 Code",
+ "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
"lifecycle": "active",
"contextWindow": 262144,
"maxOutputTokens": 262144,
- "knowledgeCutoff": "2024-08",
- "lastUpdated": "2025-11-06",
+ "knowledgeCutoff": "2025-01",
+ "structuredOutput": true,
+ "lastUpdated": "2026-06-12",
"capabilities": {
- "vision": false,
+ "vision": true,
"reasoning": true,
"functionCalling": true
},
"modalities": {
- "input": ["text"],
+ "input": ["text", "image", "video"],
"output": ["text"]
}
},
- "kimi-k2-turbo-preview": {
- "displayName": "Kimi K2 Turbo",
- "description": "Fast Kimi model for responsive chat, coding help, and agent loops",
+ "kimi-k2.7-code-highspeed": {
+ "displayName": "Kimi K2.7 Code HighSpeed",
+ "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
"lifecycle": "active",
"contextWindow": 262144,
"maxOutputTokens": 262144,
- "knowledgeCutoff": "2024-10",
- "lastUpdated": "2025-09-05",
+ "knowledgeCutoff": "2025-01",
+ "structuredOutput": true,
+ "lastUpdated": "2026-06-12",
"capabilities": {
- "vision": false,
- "reasoning": false,
+ "vision": true,
+ "reasoning": true,
"functionCalling": true
},
"modalities": {
- "input": ["text"],
+ "input": ["text", "image", "video"],
"output": ["text"]
}
},
- "kimi-k2.5": {
- "displayName": "Kimi K2.5",
- "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work",
+ "kimi-k3": {
+ "displayName": "Kimi K3",
+ "description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
"lifecycle": "active",
- "contextWindow": 262144,
- "maxOutputTokens": 262144,
- "knowledgeCutoff": "2025-01",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 131072,
"structuredOutput": true,
- "lastUpdated": "2026-01",
+ "lastUpdated": "2026-07-16",
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
+ "efforts": ["low", "high", "max"],
"toggle": true
},
"modalities": {
"input": ["text", "image", "video"],
"output": ["text"]
}
- },
+ }
+ },
+ "moonshot-global": {
"kimi-k2.6": {
"displayName": "Kimi K2.6",
"description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context",
@@ -13574,6 +13867,28 @@
"input": ["text"],
"output": ["text"]
}
+ },
+ "z-ai/glm-5.3-flash": {
+ "displayName": "GLM-5.3-Flash",
+ "description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-26",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf"],
+ "output": ["text"]
+ }
}
},
"ollama-cloud": {
@@ -13642,6 +13957,51 @@
"output": ["text"]
}
},
+ "deepseek-v4-pro:0813": {
+ "displayName": "DeepSeek V4 Pro 0813",
+ "description": "DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 1048576,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-22",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "deepseek-v4.1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"gemma4:31b": {
"displayName": "gemma4:31b",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
@@ -14397,7 +14757,7 @@
"gpt-5.2-chat-latest": {
"displayName": "GPT-5.2 Chat",
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 128000,
"maxOutputTokens": 16384,
"knowledgeCutoff": "2025-08-31",
@@ -14442,7 +14802,7 @@
"gpt-5.3-chat-latest": {
"displayName": "GPT-5.3 Chat (latest)",
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 128000,
"maxOutputTokens": 16384,
"knowledgeCutoff": "2025-08-31",
@@ -14734,6 +15094,29 @@
"output": ["text"]
}
},
+ "gpt-6-astra": {
+ "displayName": "GPT-6 Astra",
+ "description": "GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-04-30",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"gpt-image-1": {
"displayName": "gpt-image-1",
"description": "OpenAI image model for production generation, edits, and brand-safe visual workflows",
@@ -15077,6 +15460,27 @@
"output": ["text"]
}
},
+ "claude-fable-5-1": {
+ "displayName": "Claude Fable 5.1",
+ "description": "Claude model for demanding reasoning and long-horizon agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-06",
+ "lastUpdated": "2026-09-01",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"claude-haiku-4-5": {
"displayName": "Claude Haiku 4.5",
"description": "Fast Claude model for responsive assistance, classification, and lightweight agents",
@@ -15342,6 +15746,28 @@
"output": ["text"]
}
},
+ "deepseek-v4-flash-vision-exp": {
+ "displayName": "DeepSeek V4 Flash Vision Exp",
+ "description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-21",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"deepseek-v4-pro": {
"displayName": "DeepSeek V4 Pro",
"description": "Open MoE flagship with million-token context for coding and long agent runs",
@@ -15519,6 +15945,27 @@
"output": ["text"]
}
},
+ "gemini-3.8-flash": {
+ "displayName": "Gemini 3.8 Flash",
+ "description": "Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "audio", "pdf"],
+ "output": ["text"]
+ }
+ },
"glm-4.6": {
"displayName": "GLM-4.6",
"description": "Legacy model retained for compatibility with older integrations",
@@ -15668,6 +16115,48 @@
"output": ["text"]
}
},
+ "glm-5.3": {
+ "displayName": "GLM-5.3",
+ "description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-14",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "glm-5.3-flash": {
+ "displayName": "GLM-5.3-Flash",
+ "description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-26",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf"],
+ "output": ["text"]
+ }
+ },
"gpt-5": {
"displayName": "GPT-5",
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
@@ -16128,6 +16617,29 @@
"output": ["text"]
}
},
+ "gpt-6-astra": {
+ "displayName": "GPT-6 Astra",
+ "description": "GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-04-30",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"grok-4.5": {
"displayName": "Grok 4.5",
"description": "xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk",
@@ -16212,6 +16724,7 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "deprecated",
"contextWindow": 190000,
+ "inputLimit": 192000,
"maxOutputTokens": 64000,
"structuredOutput": true,
"lastUpdated": "2026-07-06",
@@ -16764,6 +17277,49 @@
"output": ["text"]
}
},
+ "muse-spark-1.3": {
+ "displayName": "Muse Spark 1.3",
+ "description": "Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf", "audio"],
+ "output": ["text"]
+ }
+ },
+ "muse-spark-1.3-contributor-free": {
+ "displayName": "Muse Spark 1.3 Free",
+ "description": "Muse Spark 1.3 is a multimodal reasoning model from Meta for coding and agentic workflows.",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf", "audio"],
+ "output": ["text"]
+ }
+ },
"nemotron-3-super-free": {
"displayName": "Nemotron 3 Super Free",
"description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads",
@@ -17054,6 +17610,28 @@
"output": ["text"]
}
},
+ "deepseek-v4.1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"glm-5": {
"displayName": "GLM-5",
"description": "Legacy model retained for compatibility with older integrations",
@@ -17133,7 +17711,7 @@
}
},
"glm-5.3-flash": {
- "displayName": "GLM-5.3-Flash (2x usage)",
+ "displayName": "GLM-5.3-Flash",
"description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
"lifecycle": "active",
"contextWindow": 1000000,
@@ -17224,7 +17802,8 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 256000,
- "maxOutputTokens": 64000,
+ "inputLimit": 192000,
+ "maxOutputTokens": 128000,
"lastUpdated": "2026-07-06",
"capabilities": {
"vision": false,
@@ -17505,6 +18084,48 @@
"output": ["text"]
}
},
+ "muse-spark-1.3-contributor": {
+ "displayName": "Muse Spark 1.3 Contributor",
+ "description": "Muse Spark 1.3 is a multimodal reasoning model from Meta for coding and agentic workflows.",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf", "audio"],
+ "output": ["text"]
+ }
+ },
+ "omen-alpha": {
+ "displayName": "Omen Alpha",
+ "description": "oH man anothEr aLPha ModEl",
+ "lifecycle": "deprecated",
+ "contextWindow": 500000,
+ "maxOutputTokens": 128000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"ox-alpha-free": {
"displayName": "Ox Alpha Free (Unlimited)",
"description": "Stealth reasoning model for coding, agentic tasks, and tool use",
@@ -17677,7 +18298,7 @@
}
},
"~anthropic/claude-haiku-latest": {
- "displayName": "Anthropic Claude Haiku Latest",
+ "displayName": "Claude Haiku Latest",
"description": "Fast Claude model for responsive assistance, classification, and lightweight agents",
"lifecycle": "active",
"contextWindow": 200000,
@@ -17720,7 +18341,7 @@
}
},
"~anthropic/claude-sonnet-latest": {
- "displayName": "Anthropic Claude Sonnet Latest",
+ "displayName": "Claude Sonnet Latest",
"description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
"lifecycle": "active",
"contextWindow": 1000000,
@@ -17742,6 +18363,50 @@
"output": ["text"]
}
},
+ "~deepseek/deepseek-flash-latest": {
+ "displayName": "DeepSeek Flash Latest",
+ "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 943718,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-14",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "~deepseek/deepseek-pro-latest": {
+ "displayName": "DeepSeek Pro Latest",
+ "description": "Flagship DeepSeek model for coding, reasoning, and agentic work",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 943718,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-14",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"~deepseek/deepseek-v4-flash-latest": {
"displayName": "DeepSeek V4 Flash Latest",
"description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
@@ -17765,7 +18430,7 @@
}
},
"~google/gemini-flash-latest": {
- "displayName": "Google Gemini Flash Latest",
+ "displayName": "Gemini Flash Latest",
"description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
"lifecycle": "active",
"contextWindow": 1048576,
@@ -17787,7 +18452,7 @@
}
},
"~google/gemini-pro-latest": {
- "displayName": "Google Gemini Pro Latest",
+ "displayName": "Gemini Pro Latest",
"description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis",
"lifecycle": "active",
"contextWindow": 1048576,
@@ -17809,7 +18474,7 @@
}
},
"~moonshotai/kimi-latest": {
- "displayName": "MoonshotAI Kimi Latest",
+ "displayName": "Kimi Latest",
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
"lifecycle": "active",
"contextWindow": 1048576,
@@ -17830,15 +18495,36 @@
"output": ["text"]
}
},
- "~openai/gpt-latest": {
- "displayName": "OpenAI GPT Latest",
+ "~openai/gpt-astra-latest": {
+ "displayName": "GPT Astra Latest",
+ "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-11",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["pdf", "image", "text"],
+ "output": ["text"]
+ }
+ },
+ "~openai/gpt-luna-latest": {
+ "displayName": "GPT Luna Latest",
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
"lifecycle": "active",
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"knowledgeCutoff": "2026-02-16",
"structuredOutput": true,
- "lastUpdated": "2026-04-27",
+ "lastUpdated": "2026-09-11",
"capabilities": {
"vision": true,
"reasoning": true,
@@ -17853,7 +18539,7 @@
}
},
"~openai/gpt-mini-latest": {
- "displayName": "OpenAI GPT Mini Latest",
+ "displayName": "GPT Mini Latest",
"description": "Compact GPT model for low-latency assistance and high-volume workloads",
"lifecycle": "active",
"contextWindow": 400000,
@@ -17874,6 +18560,50 @@
"output": ["text"]
}
},
+ "~openai/gpt-sol-latest": {
+ "displayName": "GPT Sol Latest",
+ "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-02-16",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-11",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["pdf", "image", "text"],
+ "output": ["text"]
+ }
+ },
+ "~openai/gpt-terra-latest": {
+ "displayName": "GPT Terra Latest",
+ "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-02-16",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-11",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["pdf", "image", "text"],
+ "output": ["text"]
+ }
+ },
"~x-ai/grok-latest": {
"displayName": "Grok Latest",
"description": "Grok model for agentic tool use, reasoning, coding, and live assistance",
@@ -17895,12 +18625,33 @@
"output": ["text"]
}
},
+ "~z-ai/glm-flash-latest": {
+ "displayName": "GLM Flash Latest",
+ "description": "GLM vision model for visual reasoning, documents, and multimodal agents",
+ "lifecycle": "active",
+ "contextWindow": 1310720,
+ "maxOutputTokens": 131072,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-27",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video"],
+ "output": ["text"]
+ }
+ },
"~z-ai/glm-latest": {
"displayName": "GLM Latest",
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
"lifecycle": "active",
"contextWindow": 1310720,
- "maxOutputTokens": 943718,
+ "maxOutputTokens": 131072,
"structuredOutput": true,
"lastUpdated": "2026-08-19",
"capabilities": {
@@ -18145,6 +18896,28 @@
"output": ["text"]
}
},
+ "anthropic/claude-fable-5.1": {
+ "displayName": "Claude Fable 5.1",
+ "description": "Claude model for creative writing, analysis, and controlled agent workflows",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-06",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-01",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"anthropic/claude-haiku-4.5": {
"displayName": "Claude Haiku 4.5 (latest)",
"description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat",
@@ -18279,29 +19052,6 @@
"output": ["text"]
}
},
- "anthropic/claude-opus-4.7-fast": {
- "displayName": "Claude Opus 4.7 (Fast)",
- "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
- "lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 128000,
- "knowledgeCutoff": "2026-01-31",
- "structuredOutput": true,
- "lastUpdated": "2026-04-16",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "efforts": ["low", "medium", "high", "xhigh", "max"],
- "toggle": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
"anthropic/claude-opus-4.8": {
"displayName": "Claude Opus 4.8",
"description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents",
@@ -18325,29 +19075,6 @@
"output": ["text"]
}
},
- "anthropic/claude-opus-4.8-fast": {
- "displayName": "Claude Opus 4.8 (Fast)",
- "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
- "lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 128000,
- "knowledgeCutoff": "2026-01",
- "structuredOutput": true,
- "lastUpdated": "2026-05-28",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "efforts": ["low", "medium", "high", "xhigh", "max"],
- "toggle": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
"anthropic/claude-opus-5": {
"displayName": "Claude Opus 5",
"description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
@@ -18371,29 +19098,6 @@
"output": ["text"]
}
},
- "anthropic/claude-opus-5-fast": {
- "displayName": "Claude Opus 5 (Fast)",
- "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
- "lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 128000,
- "knowledgeCutoff": "2026-05",
- "structuredOutput": true,
- "lastUpdated": "2026-07-24",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "efforts": ["low", "medium", "high", "xhigh", "max"],
- "toggle": true
- },
- "modalities": {
- "input": ["text", "image", "pdf"],
- "output": ["text"]
- }
- },
"anthropic/claude-sonnet-4": {
"displayName": "Claude Sonnet 4",
"description": "Balanced Claude model for coding, analysis, agent workflows, and cost control",
@@ -18832,7 +19536,7 @@
"description": "DeepSeek chat model for instruction following, coding, and analysis",
"lifecycle": "active",
"contextWindow": 163840,
- "maxOutputTokens": 144900,
+ "maxOutputTokens": 32768,
"knowledgeCutoff": "2025-03-31",
"structuredOutput": true,
"lastUpdated": "2025-08-21",
@@ -18856,7 +19560,7 @@
"contextWindow": 64000,
"maxOutputTokens": 16000,
"knowledgeCutoff": "2024-07",
- "structuredOutput": true,
+ "structuredOutput": false,
"lastUpdated": "2025-05-29",
"capabilities": {
"vision": false,
@@ -19026,8 +19730,8 @@
"description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
"lifecycle": "active",
"contextWindow": 1048576,
- "maxOutputTokens": 384000,
- "structuredOutput": false,
+ "maxOutputTokens": 943718,
+ "structuredOutput": true,
"lastUpdated": "2026-08-21",
"capabilities": {
"vision": true,
@@ -19088,6 +19792,29 @@
"output": ["text"]
}
},
+ "deepseek/deepseek-v4.1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"dots-studio/dots-3-note-preview:free": {
"displayName": "Dots3-Note Preview (free)",
"description": "Multimodal reasoning model for visual analysis, planning, and tool use",
@@ -19211,25 +19938,6 @@
"output": ["text"]
}
},
- "google/gemini-2.5-pro-preview-05-06": {
- "displayName": "Gemini 2.5 Pro Preview 05-06",
- "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis",
- "lifecycle": "active",
- "contextWindow": 1048576,
- "maxOutputTokens": 65535,
- "knowledgeCutoff": "2025-01-31",
- "structuredOutput": true,
- "lastUpdated": "2025-05-07",
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text", "image", "pdf", "audio", "video"],
- "output": ["text"]
- }
- },
"google/gemini-3-flash-preview": {
"displayName": "Gemini 3 Flash Preview",
"description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs",
@@ -19273,7 +19981,7 @@
}
},
"google/gemini-3-pro-image-preview": {
- "displayName": "Nano Banana Pro",
+ "displayName": "Nano Banana Pro Preview",
"description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits",
"lifecycle": "active",
"contextWindow": 65536,
@@ -19315,7 +20023,7 @@
}
},
"google/gemini-3.1-flash-image-preview": {
- "displayName": "Nano Banana 2",
+ "displayName": "Nano Banana 2 Preview",
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
"lifecycle": "active",
"contextWindow": 65536,
@@ -19538,6 +20246,27 @@
"output": ["text"]
}
},
+ "google/gemini-3.8-flash": {
+ "displayName": "Gemini 3.8 Flash",
+ "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "audio", "pdf"],
+ "output": ["text"]
+ }
+ },
"google/gemma-2-27b-it": {
"displayName": "Gemma 2 27B",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
@@ -19558,14 +20287,14 @@
}
},
"google/gemma-3-12b-it": {
- "displayName": "Gemma 3 12B",
+ "displayName": "Gemma 3 12B IT",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
"lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 16384,
- "knowledgeCutoff": "2024-08-31",
+ "knowledgeCutoff": "2024-08",
"structuredOutput": true,
- "lastUpdated": "2025-03-13",
+ "lastUpdated": "2025-03-12",
"capabilities": {
"vision": true,
"reasoning": false,
@@ -19577,12 +20306,12 @@
}
},
"google/gemma-3-27b-it": {
- "displayName": "Gemma 3 27B",
+ "displayName": "Gemma 3 27B IT",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
"lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 117964,
- "knowledgeCutoff": "2024-08-31",
+ "knowledgeCutoff": "2024-08",
"structuredOutput": true,
"lastUpdated": "2025-03-12",
"capabilities": {
@@ -19596,14 +20325,14 @@
}
},
"google/gemma-3-4b-it": {
- "displayName": "Gemma 3 4B",
+ "displayName": "Gemma 3 4B IT",
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
"lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 16384,
- "knowledgeCutoff": "2024-08-31",
+ "knowledgeCutoff": "2024-08",
"structuredOutput": true,
- "lastUpdated": "2025-03-13",
+ "lastUpdated": "2025-03-12",
"capabilities": {
"vision": true,
"reasoning": false,
@@ -19619,7 +20348,7 @@
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 16384,
+ "maxOutputTokens": 235929,
"structuredOutput": true,
"lastUpdated": "2026-04-02",
"capabilities": {
@@ -19775,24 +20504,6 @@
"output": ["text"]
}
},
- "ibm-granite/granite-4.1-8b": {
- "displayName": "Granite 4.1 8B",
- "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
- "lifecycle": "active",
- "contextWindow": 131072,
- "maxOutputTokens": 117964,
- "structuredOutput": true,
- "lastUpdated": "2026-04-30",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"ibm-granite/granite-4.2-8b": {
"displayName": "Granite 4.2 8B",
"description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use",
@@ -19835,8 +20546,29 @@
"output": ["text"]
}
},
+ "inception/mercury-2.5": {
+ "displayName": "Mercury 2.5",
+ "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 260000,
+ "maxOutputTokens": 65536,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-08",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"inclusionai/ling-3.0-flash": {
- "displayName": "Ling-3.0-flash",
+ "displayName": "Ling 3.0 Flash",
"description": "Efficient model for low-latency assistance, extraction, and routine automation",
"lifecycle": "active",
"contextWindow": 262144,
@@ -19856,6 +20588,27 @@
"output": ["text"]
}
},
+ "inclusionai/ling-3.0-flash-fin": {
+ "displayName": "Ling 3.0 Flash Fin",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 235929,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-27",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"inclusionai/ling-3.0-flash-fin:free": {
"displayName": "Ling 3.0 Flash Fin (free)",
"description": "Efficient model for low-latency assistance, extraction, and routine automation",
@@ -19878,6 +20631,107 @@
"output": ["text"]
}
},
+ "inclusionai/ling-3.0-flash-sante:free": {
+ "displayName": "Ling 3.0 Flash Sante (free)",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 32768,
+ "structuredOutput": false,
+ "lastUpdated": "2026-09-04",
+ "isFree": true,
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "inclusionai/ling-3.0-flash-vl": {
+ "displayName": "Ling 3.0 Flash VL",
+ "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 32768,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image", "video"],
+ "output": ["text"]
+ }
+ },
+ "inclusionai/ling-3.0-flash-vl:free": {
+ "displayName": "Ling 3.0 Flash VL (free)",
+ "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 32768,
+ "structuredOutput": false,
+ "lastUpdated": "2026-09-10",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image", "video"],
+ "output": ["text"]
+ }
+ },
+ "inference-net/schematron-v2-small": {
+ "displayName": "Schematron V2 Small",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-12",
+ "capabilities": {
+ "vision": false,
+ "reasoning": false,
+ "functionCalling": false
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "inference-net/schematron-v2-turbo": {
+ "displayName": "Schematron V2 Turbo",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 128000,
+ "maxOutputTokens": 8192,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-12",
+ "capabilities": {
+ "vision": false,
+ "reasoning": false,
+ "functionCalling": false
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"kwaipilot/kat-coder-pro-v2": {
"displayName": "KAT-Coder-Pro V2",
"description": "Coding model for repository understanding, refactors, and agentic engineering tasks",
@@ -19974,12 +20828,12 @@
}
},
"meta-llama/llama-3.1-70b-instruct": {
- "displayName": "Llama 3.1 70B Instruct",
+ "displayName": "Llama-3.1-70B-Instruct",
"description": "Open Llama instruction model for multilingual chat, reasoning, and coding",
"lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 16384,
- "knowledgeCutoff": "2023-12-31",
+ "knowledgeCutoff": "2023-12",
"structuredOutput": true,
"lastUpdated": "2024-07-23",
"capabilities": {
@@ -20054,7 +20908,7 @@
"description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting",
"lifecycle": "active",
"contextWindow": 131072,
- "maxOutputTokens": 115200,
+ "maxOutputTokens": 16384,
"knowledgeCutoff": "2023-12",
"structuredOutput": true,
"lastUpdated": "2024-12-06",
@@ -20073,7 +20927,7 @@
"description": "Open multimodal Llama model for strong reasoning and fast responses",
"lifecycle": "active",
"contextWindow": 1048576,
- "maxOutputTokens": 115200,
+ "maxOutputTokens": 16384,
"knowledgeCutoff": "2024-08-31",
"structuredOutput": true,
"lastUpdated": "2025-04-05",
@@ -20130,7 +20984,7 @@
"description": "Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.",
"lifecycle": "active",
"contextWindow": 131072,
- "maxOutputTokens": 16384,
+ "maxOutputTokens": 117964,
"knowledgeCutoff": "2026-01-04",
"structuredOutput": true,
"lastUpdated": "2026-08-10",
@@ -20210,6 +21064,48 @@
"output": ["text"]
}
},
+ "meta/muse-spark-1.3": {
+ "displayName": "Muse Spark 1.3",
+ "description": "Open Llama multimodal model for image understanding and text reasoning",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 943718,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf", "audio"],
+ "output": ["text"]
+ }
+ },
+ "meta/muse-spark-1.3-contributor": {
+ "displayName": "Muse Spark 1.3 Contributor",
+ "description": "Open Llama multimodal model for image understanding and text reasoning",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 943718,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "video", "pdf", "audio"],
+ "output": ["text"]
+ }
+ },
"microsoft/phi-4": {
"displayName": "Phi 4",
"description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
@@ -20379,25 +21275,6 @@
"output": ["text"]
}
},
- "minimax/minimax-m2.7:free": {
- "displayName": "MiniMax M2.7 (free)",
- "description": "MiniMax model for chat, coding, office work, and agentic tasks",
- "lifecycle": "active",
- "contextWindow": 196608,
- "maxOutputTokens": 176947,
- "structuredOutput": false,
- "lastUpdated": "2026-03-18",
- "isFree": true,
- "capabilities": {
- "vision": false,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"minimax/minimax-m3": {
"displayName": "MiniMax-M3",
"description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
@@ -20419,28 +21296,6 @@
"output": ["text"]
}
},
- "minimax/minimax-m3:free": {
- "displayName": "MiniMax M3 (free)",
- "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks",
- "lifecycle": "active",
- "contextWindow": 1048576,
- "maxOutputTokens": 943718,
- "structuredOutput": false,
- "lastUpdated": "2026-06-01",
- "isFree": true,
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "toggle": true
- },
- "modalities": {
- "input": ["text", "image", "video"],
- "output": ["text"]
- }
- },
"mistralai/codestral-2508": {
"displayName": "Codestral 2508",
"description": "Mistral coding model for code completion, generation, and developer workflows",
@@ -20751,7 +21606,7 @@
"displayName": "Mistral Small 3.2 24B",
"description": "Efficient Mistral model for fast chat, extraction, and production assistants",
"lifecycle": "active",
- "contextWindow": 131072,
+ "contextWindow": 256000,
"maxOutputTokens": 16384,
"knowledgeCutoff": "2023-10-31",
"structuredOutput": true,
@@ -20792,7 +21647,7 @@
"contextWindow": 32768,
"maxOutputTokens": 26214,
"structuredOutput": true,
- "lastUpdated": "2025-10-30",
+ "lastUpdated": "2025-07-15",
"capabilities": {
"vision": false,
"reasoning": false,
@@ -20808,7 +21663,7 @@
"description": "Kimi model for long-context chat, coding, and agentic reasoning",
"lifecycle": "active",
"contextWindow": 131072,
- "maxOutputTokens": 100352,
+ "maxOutputTokens": 98304,
"knowledgeCutoff": "2024-12-31",
"structuredOutput": false,
"lastUpdated": "2025-07-11",
@@ -20827,7 +21682,7 @@
"description": "Kimi model for long-context chat, coding, and agentic reasoning",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 100352,
+ "maxOutputTokens": 98304,
"knowledgeCutoff": "2024-12-31",
"structuredOutput": true,
"lastUpdated": "2025-09-04",
@@ -20846,7 +21701,7 @@
"description": "Thinking Kimi model for slower research passes, planning, and hard technical questions",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 100352,
+ "maxOutputTokens": 98304,
"knowledgeCutoff": "2024-08",
"structuredOutput": true,
"lastUpdated": "2025-11-06",
@@ -20981,42 +21836,44 @@
"output": ["text"]
}
},
- "nex-agi/nex-n2-mini": {
- "displayName": "Nex-N2-Mini",
- "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "nex-agi/nex-n2.5-mini:free": {
+ "displayName": "Nex-N2.5-Mini (free)",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
"lifecycle": "active",
"contextWindow": 262144,
"maxOutputTokens": 235929,
"structuredOutput": true,
- "lastUpdated": "2026-06-24",
+ "lastUpdated": "2026-09-08",
+ "isFree": true,
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "toggle": true
+ "efforts": ["none", "medium", "high"]
},
"modalities": {
"input": ["text", "image"],
"output": ["text"]
}
},
- "nex-agi/nex-n2-pro": {
- "displayName": "Nex-N2-Pro",
+ "nex-agi/nex-n2.5-pro:free": {
+ "displayName": "Nex-N2.5-Pro (free)",
"description": "Multimodal reasoning model for visual analysis, planning, and tool use",
"lifecycle": "active",
"contextWindow": 262144,
"maxOutputTokens": 235929,
- "structuredOutput": false,
- "lastUpdated": "2026-06-08",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-08",
+ "isFree": true,
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "toggle": true
+ "efforts": ["none", "medium", "high"]
},
"modalities": {
"input": ["text", "image"],
@@ -21083,34 +21940,12 @@
"output": ["text"]
}
},
- "nousresearch/hermes-4-70b": {
- "displayName": "Hermes 4 70B",
- "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use",
- "lifecycle": "active",
- "contextWindow": 131072,
- "maxOutputTokens": 117964,
- "knowledgeCutoff": "2024-08-31",
- "structuredOutput": false,
- "lastUpdated": "2025-08-26",
- "capabilities": {
- "vision": false,
- "reasoning": true,
- "functionCalling": false
- },
- "thinkingOptions": {
- "toggle": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"nvidia/nemotron-3-nano-30b-a3b": {
"displayName": "Nemotron 3 Nano 30B A3B",
"description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 228000,
+ "maxOutputTokens": 235929,
"structuredOutput": true,
"lastUpdated": "2025-12-15",
"capabilities": {
@@ -21152,8 +21987,8 @@
"displayName": "Nemotron 3 Super 120B A12B",
"description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads",
"lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 16384,
+ "contextWindow": 262144,
+ "maxOutputTokens": 235929,
"structuredOutput": true,
"lastUpdated": "2026-03-11",
"capabilities": {
@@ -21198,7 +22033,7 @@
"description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 16384,
+ "maxOutputTokens": 182520,
"structuredOutput": true,
"lastUpdated": "2026-06-04",
"capabilities": {
@@ -21238,6 +22073,27 @@
"output": ["text"]
}
},
+ "nvidia/nemotron-3.5-content-safety": {
+ "displayName": "Nemotron 3.5 Content Safety",
+ "description": "Safety model for policy screening, moderation, and risk-aware routing workflows",
+ "lifecycle": "active",
+ "contextWindow": 131072,
+ "maxOutputTokens": 117964,
+ "structuredOutput": false,
+ "lastUpdated": "2026-06-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": false
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"nvidia/nemotron-3.5-content-safety:free": {
"displayName": "Nemotron 3.5 Content Safety (free)",
"description": "Safety model for policy screening, moderation, and risk-aware routing workflows",
@@ -21417,25 +22273,6 @@
"output": ["text"]
}
},
- "openai/gpt-4-turbo-preview": {
- "displayName": "GPT-4 Turbo Preview",
- "description": "Compact GPT model for low-latency assistance and high-volume workloads",
- "lifecycle": "active",
- "contextWindow": 128000,
- "maxOutputTokens": 4096,
- "knowledgeCutoff": "2023-12-31",
- "structuredOutput": true,
- "lastUpdated": "2024-01-25",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"openai/gpt-4.1": {
"displayName": "GPT-4.1",
"description": "Long-lived GPT workhorse for coding, instruction following, and production apps",
@@ -22237,6 +23074,50 @@
"output": ["text"]
}
},
+ "openai/gpt-6-astra": {
+ "displayName": "GPT-6 Astra",
+ "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-04-30",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
+ "openai/gpt-6-astra-pro": {
+ "displayName": "GPT-6 Astra Pro",
+ "description": "Frontier GPT model for professional reasoning, coding, and multimodal work",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["pdf", "image", "text"],
+ "output": ["text"]
+ }
+ },
"openai/gpt-audio": {
"displayName": "GPT Audio",
"description": "Speech generation model for controllable voice, narration, and audio delivery",
@@ -22334,7 +23215,7 @@
}
},
"openai/gpt-oss-safeguard-20b": {
- "displayName": "gpt-oss-safeguard-20b",
+ "displayName": "GPT OSS Safeguard 20B",
"description": "Safety model for policy screening, moderation, and risk-aware routing workflows",
"lifecycle": "active",
"contextWindow": 131072,
@@ -22925,7 +23806,7 @@
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
"contextWindow": 128000,
- "maxOutputTokens": 28800,
+ "maxOutputTokens": 115200,
"knowledgeCutoff": "2024-06-30",
"structuredOutput": true,
"lastUpdated": "2025-02-01",
@@ -23027,7 +23908,7 @@
"lifecycle": "active",
"contextWindow": 131072,
"maxOutputTokens": 16384,
- "structuredOutput": false,
+ "structuredOutput": true,
"lastUpdated": "2025-04-28",
"capabilities": {
"vision": false,
@@ -23264,7 +24145,7 @@
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 235929,
+ "maxOutputTokens": 16384,
"knowledgeCutoff": "2025-04",
"structuredOutput": true,
"lastUpdated": "2025-09",
@@ -23432,7 +24313,7 @@
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 81920,
+ "maxOutputTokens": 65536,
"structuredOutput": true,
"lastUpdated": "2026-02-23",
"capabilities": {
@@ -23474,7 +24355,7 @@
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 235929,
+ "maxOutputTokens": 65536,
"structuredOutput": true,
"lastUpdated": "2026-02-23",
"capabilities": {
@@ -23495,7 +24376,7 @@
"description": "Large open Qwen multimodal MoE for visual agents and long technical tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 65536,
+ "maxOutputTokens": 235929,
"structuredOutput": true,
"lastUpdated": "2026-02-15",
"capabilities": {
@@ -23601,7 +24482,7 @@
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
- "maxOutputTokens": 235929,
+ "maxOutputTokens": 65536,
"structuredOutput": true,
"lastUpdated": "2026-04-22",
"capabilities": {
@@ -23773,7 +24654,7 @@
"description": "Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows",
"lifecycle": "active",
"contextWindow": 1048576,
- "maxOutputTokens": 262144,
+ "maxOutputTokens": 131072,
"structuredOutput": true,
"lastUpdated": "2026-08-12",
"capabilities": {
@@ -23832,14 +24713,14 @@
"output": ["text"]
}
},
- "qwen/qwen3.8-max": {
- "displayName": "Qwen3.8 Max",
+ "qwen/qwen3.8-max-0902": {
+ "displayName": "Qwen3.8 Max 0902",
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
"structuredOutput": true,
- "lastUpdated": "2026-08-03",
+ "lastUpdated": "2026-09-02",
"capabilities": {
"vision": true,
"reasoning": true,
@@ -23926,6 +24807,27 @@
"output": ["text"]
}
},
+ "sakana/fugu-max": {
+ "displayName": "Fugu Max",
+ "description": "Multi-agent model for routing expert agents across complex analytical tasks",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-11",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"sakana/fugu-ultra": {
"displayName": "Fugu Ultra",
"description": "Quality-first multi-agent model for hard research, analysis, and competitions",
@@ -23947,6 +24849,28 @@
"output": ["text"]
}
},
+ "sakana/fugu-ultra-v2": {
+ "displayName": "Fugu Ultra v2",
+ "description": "Quality-first multi-agent model for hard research, analysis, and competitions",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-08-28",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-11",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"sakana/sakana-namazu": {
"displayName": "Sakana Namazu",
"description": "Multi-agent model for routing expert agents across complex analytical tasks",
@@ -24148,6 +25072,7 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 262144,
+ "inputLimit": 192000,
"maxOutputTokens": 128000,
"structuredOutput": true,
"lastUpdated": "2026-07-06",
@@ -24249,7 +25174,7 @@
"description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
"lifecycle": "active",
"contextWindow": 1024000,
- "maxOutputTokens": 26214,
+ "maxOutputTokens": 819200,
"knowledgeCutoff": "2024-04-30",
"structuredOutput": true,
"lastUpdated": "2024-11-08",
@@ -24354,7 +25279,7 @@
"description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
"lifecycle": "active",
"contextWindow": 6144,
- "maxOutputTokens": 4096,
+ "maxOutputTokens": 5529,
"knowledgeCutoff": "2023-06-30",
"structuredOutput": true,
"lastUpdated": "2023-07-22",
@@ -24734,8 +25659,8 @@
"displayName": "GLM-4.7-Flash",
"description": "Budget GLM lane for fast coding help, routing, and everyday automation",
"lifecycle": "active",
- "contextWindow": 202752,
- "maxOutputTokens": 16384,
+ "contextWindow": 200000,
+ "maxOutputTokens": 117964,
"knowledgeCutoff": "2025-04",
"structuredOutput": true,
"lastUpdated": "2026-01-19",
@@ -24820,32 +25745,9 @@
"description": "Open flagship GLM for long-horizon coding agents and million-token context work",
"lifecycle": "active",
"contextWindow": 1048576,
- "maxOutputTokens": 262144,
- "structuredOutput": true,
- "lastUpdated": "2026-06-13",
- "capabilities": {
- "vision": false,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "efforts": ["high", "xhigh"],
- "toggle": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
- "z-ai/glm-5.2:free": {
- "displayName": "GLM 5.2 (free)",
- "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
- "lifecycle": "active",
- "contextWindow": 256000,
- "maxOutputTokens": 230400,
+ "maxOutputTokens": 128000,
"structuredOutput": true,
"lastUpdated": "2026-06-13",
- "isFree": true,
"capabilities": {
"vision": false,
"reasoning": true,
@@ -24865,7 +25767,7 @@
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
"lifecycle": "active",
"contextWindow": 1310720,
- "maxOutputTokens": 131072,
+ "maxOutputTokens": 943717,
"structuredOutput": true,
"lastUpdated": "2026-08-14",
"capabilities": {
@@ -26452,6 +27354,28 @@
"output": ["text"]
}
},
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"essentialai/Rnj-1-Instruct": {
"displayName": "Rnj-1 Instruct",
"description": "Open-weight instruction model for adaptable chat and self-hosted production workloads",
@@ -27212,7 +28136,8 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 256000,
- "maxOutputTokens": 64000,
+ "inputLimit": 192000,
+ "maxOutputTokens": 128000,
"lastUpdated": "2026-07-06",
"isFree": true,
"capabilities": {
@@ -27257,7 +28182,8 @@
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
"lifecycle": "active",
"contextWindow": 256000,
- "maxOutputTokens": 64000,
+ "inputLimit": 192000,
+ "maxOutputTokens": 128000,
"lastUpdated": "2026-07-06",
"isFree": true,
"capabilities": {
@@ -27913,14 +28839,13 @@
"output": ["text"]
}
},
- "alibaba/qwen3.8-flash-next": {
- "displayName": "Qwen 3.8 Flash Next",
- "description": "Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding",
+ "alibaba/qwen3.8-max": {
+ "displayName": "Qwen 3.8 Max",
+ "description": "Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows",
"lifecycle": "active",
- "contextWindow": 1048576,
- "maxOutputTokens": 1048576,
- "structuredOutput": true,
- "lastUpdated": "2026-08-27",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "lastUpdated": "2026-07-19",
"capabilities": {
"vision": true,
"reasoning": true,
@@ -27934,23 +28859,24 @@
"output": ["text"]
}
},
- "alibaba/qwen3.8-max": {
- "displayName": "Qwen 3.8 Max",
- "description": "Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows",
+ "alibaba/qwen3.8-max-0902": {
+ "displayName": "Qwen3.8 Max 0902",
+ "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
"lifecycle": "active",
- "contextWindow": 1000000,
+ "contextWindow": 991000,
"maxOutputTokens": 128000,
- "lastUpdated": "2026-07-19",
+ "lastUpdated": "2026-09-02",
"capabilities": {
"vision": true,
"reasoning": true,
"functionCalling": true
},
"thinkingOptions": {
- "efforts": ["low", "medium", "xhigh"]
+ "efforts": ["low", "medium", "xhigh"],
+ "toggle": true
},
"modalities": {
- "input": ["text", "image"],
+ "input": ["text", "image", "pdf"],
"output": ["text"]
}
},
@@ -28151,7 +29077,7 @@
"description": "Efficient model for low-latency assistance, extraction, and routine automation",
"lifecycle": "active",
"contextWindow": 300000,
- "maxOutputTokens": 8192,
+ "maxOutputTokens": 10000,
"knowledgeCutoff": "2024-10",
"lastUpdated": "2024-12-03",
"capabilities": {
@@ -28160,7 +29086,7 @@
"functionCalling": true
},
"modalities": {
- "input": ["text", "image", "video"],
+ "input": ["text", "image", "video", "pdf"],
"output": ["text"]
}
},
@@ -28169,7 +29095,7 @@
"description": "Efficient model for low-latency assistance, extraction, and routine automation",
"lifecycle": "active",
"contextWindow": 128000,
- "maxOutputTokens": 8192,
+ "maxOutputTokens": 10000,
"knowledgeCutoff": "2024-10",
"lastUpdated": "2024-12-03",
"capabilities": {
@@ -28187,7 +29113,7 @@
"description": "Flagship model for demanding analysis, coding, and production agent workflows",
"lifecycle": "active",
"contextWindow": 300000,
- "maxOutputTokens": 8192,
+ "maxOutputTokens": 10000,
"knowledgeCutoff": "2024-10",
"lastUpdated": "2024-12-03",
"capabilities": {
@@ -28196,7 +29122,7 @@
"functionCalling": true
},
"modalities": {
- "input": ["text", "image", "video"],
+ "input": ["text", "image", "video", "pdf"],
"output": ["text"]
}
},
@@ -28257,6 +29183,27 @@
"output": ["text"]
}
},
+ "anthropic/claude-fable-5.1": {
+ "displayName": "Claude Fable 5.1",
+ "description": "Claude model for creative writing, analysis, and controlled agent workflows",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-06",
+ "lastUpdated": "2026-09-01",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"anthropic/claude-haiku-4.5": {
"displayName": "Claude Haiku 4.5",
"description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat",
@@ -28775,6 +29722,26 @@
"output": ["text"]
}
},
+ "bytedance/seed-2.1-turbo": {
+ "displayName": "Seed 2.1 Turbo",
+ "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 262144,
+ "maxOutputTokens": 262144,
+ "lastUpdated": "2026-06-23",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "minimal", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"bytedance/seedance-2.0": {
"displayName": "Seedance 2.0",
"description": "Video model for prompt-guided generation, editing, and motion workflows",
@@ -29050,7 +30017,7 @@
},
"deepseek/deepseek-r1": {
"displayName": "DeepSeek-R1",
- "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools",
+ "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving",
"lifecycle": "active",
"contextWindow": 128000,
"maxOutputTokens": 32768,
@@ -29059,24 +30026,7 @@
"capabilities": {
"vision": false,
"reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
- "deepseek/deepseek-v3": {
- "displayName": "DeepSeek V3 0324",
- "description": "DeepSeek chat model for instruction following, coding, and analysis",
- "lifecycle": "active",
- "contextWindow": 163840,
- "maxOutputTokens": 163840,
- "lastUpdated": "2024-12-26",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": true
+ "functionCalling": false
},
"modalities": {
"input": ["text"],
@@ -29207,8 +30157,8 @@
"displayName": "DeepSeek V4 Flash Vision Exp",
"description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
"lifecycle": "active",
- "contextWindow": 1000000,
- "maxOutputTokens": 384000,
+ "contextWindow": 1048576,
+ "maxOutputTokens": 1048576,
"structuredOutput": true,
"lastUpdated": "2026-08-21",
"capabilities": {
@@ -29270,6 +30220,29 @@
"output": ["text"]
}
},
+ "deepseek/deepseek-v4.1-flash": {
+ "displayName": "DeepSeek V4.1 Flash",
+ "description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 384000,
+ "knowledgeCutoff": "2025-05",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "xhigh"],
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"fish-audio/s1": {
"displayName": "S1",
"description": "Speech generation model for controllable voice, narration, and audio delivery",
@@ -29753,6 +30726,24 @@
"output": ["text"]
}
},
+ "google/gemini-3.8-flash": {
+ "displayName": "Gemini 3.8 Flash",
+ "description": "Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 65536,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"google/gemini-embedding-001": {
"displayName": "Gemini Embedding 001",
"description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
@@ -29981,6 +30972,26 @@
"output": ["text"]
}
},
+ "inception/mercury-2.5": {
+ "displayName": "Mercury 2.5",
+ "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 260000,
+ "maxOutputTokens": 65536,
+ "lastUpdated": "2026-09-08",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"inception/mercury-coder-small": {
"displayName": "Mercury Coder Small Beta",
"description": "Coding model for repository understanding, refactors, and agentic engineering tasks",
@@ -30057,6 +31068,84 @@
"output": ["text"]
}
},
+ "inclusionai/ling-3.0-flash-sante": {
+ "displayName": "Ling 3.0 Flash Sante",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 256000,
+ "maxOutputTokens": 32000,
+ "lastUpdated": "2026-09-04",
+ "isFree": true,
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "inclusionai/ling-3.0-flash-sante-free": {
+ "displayName": "Ling 3.0 Flash Sante (Free)",
+ "description": "Efficient model for low-latency assistance, extraction, and routine automation",
+ "lifecycle": "active",
+ "contextWindow": 256000,
+ "maxOutputTokens": 32000,
+ "lastUpdated": "2026-09-04",
+ "isFree": true,
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
+ "inclusionai/ling-3.0-flash-vl": {
+ "displayName": "Ling 3.0 Flash VL",
+ "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 256000,
+ "maxOutputTokens": 32000,
+ "lastUpdated": "2026-09-08",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
+ "inclusionai/ling-3.0-flash-vl-free": {
+ "displayName": "Ling 3.0 Flash VL (Free)",
+ "description": "Multimodal reasoning model for visual analysis, planning, and tool use",
+ "lifecycle": "active",
+ "contextWindow": 256000,
+ "maxOutputTokens": 32000,
+ "lastUpdated": "2026-09-08",
+ "isFree": true,
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "toggle": true
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"interfaze/interfaze-beta": {
"displayName": "Interfaze Beta",
"description": "Multimodal reasoning model for visual analysis, planning, and tool use",
@@ -30465,6 +31554,50 @@
"reasoning": true,
"functionCalling": true
},
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
+ "meta/muse-spark-1.3": {
+ "displayName": "Muse Spark 1.3",
+ "description": "Open Llama multimodal model for image understanding and text reasoning",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 1048576,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
+ "meta/muse-spark-1.3-contributor": {
+ "displayName": "Muse Spark 1.3 Contributor",
+ "description": "Open Llama multimodal model for image understanding and text reasoning",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 1048576,
+ "lastUpdated": "2026-09-02",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["minimal", "low", "medium", "high", "xhigh"]
+ },
"modalities": {
"input": ["text", "image", "pdf"],
"output": ["text"]
@@ -30608,24 +31741,6 @@
"output": ["text"]
}
},
- "minimax/minimax-m2.7-free": {
- "displayName": "Minimax M2.7 (Free)",
- "description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
- "lifecycle": "active",
- "contextWindow": 196608,
- "maxOutputTokens": 196608,
- "lastUpdated": "2026-03-18",
- "isFree": true,
- "capabilities": {
- "vision": false,
- "reasoning": true,
- "functionCalling": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"minimax/minimax-m2.7-highspeed": {
"displayName": "MiniMax M2.7 High Speed",
"description": "Low-latency M2.7 variant for interactive coding plans and agent loops",
@@ -30663,27 +31778,6 @@
"output": ["text"]
}
},
- "minimax/minimax-m3-free": {
- "displayName": "MiniMax M3 (Free)",
- "description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
- "lifecycle": "active",
- "contextWindow": 1048576,
- "maxOutputTokens": 1048576,
- "lastUpdated": "2026-06-01",
- "isFree": true,
- "capabilities": {
- "vision": true,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "toggle": true
- },
- "modalities": {
- "input": ["text", "image"],
- "output": ["text"]
- }
- },
"mistral/codestral": {
"displayName": "Codestral (latest)",
"description": "Mistral code model for completions, refactors, and developer IDE workflows",
@@ -32268,6 +33362,51 @@
"output": ["text"]
}
},
+ "openai/gpt-6-astra": {
+ "displayName": "GPT-6 Astra",
+ "description": "GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "knowledgeCutoff": "2026-04-30",
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
+ "openai/gpt-6-astra-fast": {
+ "displayName": "GPT-6 Astra (Fast)",
+ "description": "Fast variant of GPT-6 Astra for low-latency assistance and high-volume workloads.",
+ "lifecycle": "active",
+ "contextWindow": 1050000,
+ "inputLimit": 922000,
+ "maxOutputTokens": 128000,
+ "structuredOutput": true,
+ "lastUpdated": "2026-09-04",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["none", "low", "medium", "high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image", "pdf"],
+ "output": ["text"]
+ }
+ },
"openai/gpt-image-1": {
"displayName": "GPT Image 1",
"description": "OpenAI image model for production generation, edits, and brand-safe visual workflows",
@@ -32336,6 +33475,40 @@
"output": ["image"]
}
},
+ "openai/gpt-image-2.5-flare": {
+ "displayName": "GPT Image 2.5 Flare",
+ "description": "Image model for prompt-driven generation, editing, and visual design workflows",
+ "lifecycle": "active",
+ "contextWindow": 0,
+ "maxOutputTokens": 0,
+ "lastUpdated": "2026-09-08",
+ "capabilities": {
+ "vision": false,
+ "reasoning": false,
+ "functionCalling": false
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["image"]
+ }
+ },
+ "openai/gpt-image-2.5-sunburst": {
+ "displayName": "GPT Image 2.5 Sunburst",
+ "description": "Image model for prompt-driven generation, editing, and visual design workflows",
+ "lifecycle": "active",
+ "contextWindow": 0,
+ "maxOutputTokens": 0,
+ "lastUpdated": "2026-09-08",
+ "capabilities": {
+ "vision": false,
+ "reasoning": false,
+ "functionCalling": false
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["image"]
+ }
+ },
"openai/gpt-oss-120b": {
"displayName": "GPT OSS 120B",
"description": "Open GPT reasoning model for self-hosted agents and controllable deployments",
@@ -32410,8 +33583,8 @@
"contextWindow": 128000,
"inputLimit": 112000,
"maxOutputTokens": 16000,
- "knowledgeCutoff": "2024-10",
- "lastUpdated": "2024-12-01",
+ "structuredOutput": true,
+ "lastUpdated": "2025-10-29",
"capabilities": {
"vision": false,
"reasoning": true,
@@ -33082,6 +34255,26 @@
"output": ["image"]
}
},
+ "sakana/fugu-max": {
+ "displayName": "Fugu Max",
+ "description": "Multi-agent model for routing expert agents across complex analytical tasks",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 1000000,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"sakana/fugu-ultra": {
"displayName": "Fugu Ultra",
"description": "Quality-first multi-agent model for hard research, analysis, and competitions",
@@ -33100,6 +34293,26 @@
"output": ["text"]
}
},
+ "sakana/fugu-ultra-v2": {
+ "displayName": "Fugu Ultra v2",
+ "description": "Quality-first multi-agent model for hard research, analysis, and competitions",
+ "lifecycle": "active",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 1000000,
+ "lastUpdated": "2026-09-10",
+ "capabilities": {
+ "vision": true,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["high", "xhigh", "max"]
+ },
+ "modalities": {
+ "input": ["text", "image"],
+ "output": ["text"]
+ }
+ },
"sakana/namazu": {
"displayName": "Sakana Namazu",
"description": "Multi-agent model for routing expert agents across complex analytical tasks",
@@ -33140,6 +34353,7 @@
"lifecycle": "active",
"contextWindow": 1000000,
"maxOutputTokens": 1000000,
+ "structuredOutput": true,
"lastUpdated": "2025-11-19",
"capabilities": {
"vision": true,
@@ -33403,23 +34617,6 @@
"output": ["video"]
}
},
- "spacexai/grok-imagine-video-1.5-preview": {
- "displayName": "Grok Imagine Video 1.5 Preview",
- "description": "Image model for prompt-driven generation, editing, and visual design workflows",
- "lifecycle": "active",
- "contextWindow": 0,
- "maxOutputTokens": 0,
- "lastUpdated": "2026-05-30",
- "capabilities": {
- "vision": false,
- "reasoning": false,
- "functionCalling": false
- },
- "modalities": {
- "input": ["text"],
- "output": ["video"]
- }
- },
"spacexai/grok-stt": {
"displayName": "Grok STT",
"description": "Speech transcription model for accurate audio-to-text and captioning workflows",
@@ -33908,27 +35105,6 @@
"output": ["text"]
}
},
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
- "displayName": "MiMo V2.5 Pro UltraSpeed",
- "description": "MiMo pro model for strong multimodal reasoning and agent execution",
- "lifecycle": "active",
- "contextWindow": 1048576,
- "maxOutputTokens": 131072,
- "knowledgeCutoff": "2024-12",
- "lastUpdated": "2026-06-09",
- "capabilities": {
- "vision": false,
- "reasoning": true,
- "functionCalling": true
- },
- "thinkingOptions": {
- "toggle": true
- },
- "modalities": {
- "input": ["text"],
- "output": ["text"]
- }
- },
"zai/glm-4.5": {
"displayName": "GLM 4.5",
"description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful",
@@ -34203,6 +35379,27 @@
"output": ["text"]
}
},
+ "zai/glm-5.3-fast": {
+ "displayName": "GLM 5.3 Fast",
+ "description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
+ "lifecycle": "active",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 262144,
+ "structuredOutput": true,
+ "lastUpdated": "2026-08-14",
+ "capabilities": {
+ "vision": false,
+ "reasoning": true,
+ "functionCalling": true
+ },
+ "thinkingOptions": {
+ "efforts": ["low", "high", "max"]
+ },
+ "modalities": {
+ "input": ["text"],
+ "output": ["text"]
+ }
+ },
"zai/glm-5.3-flash": {
"displayName": "GLM 5.3 Flash",
"description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
@@ -34389,7 +35586,7 @@
"displayName": "Grok Imagine Image",
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
"lifecycle": "active",
- "contextWindow": 8000,
+ "contextWindow": 16000,
"maxOutputTokens": 0,
"lastUpdated": "2026-01-28",
"capabilities": {
@@ -34406,7 +35603,7 @@
"displayName": "Grok Imagine Image 2.0",
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
"lifecycle": "active",
- "contextWindow": 8000,
+ "contextWindow": 64000,
"maxOutputTokens": 0,
"lastUpdated": "2026-08-07",
"capabilities": {
@@ -34423,7 +35620,7 @@
"displayName": "Grok Imagine Image Quality",
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
"lifecycle": "active",
- "contextWindow": 8000,
+ "contextWindow": 16000,
"maxOutputTokens": 0,
"lastUpdated": "2026-04-03",
"capabilities": {
@@ -36556,7 +37753,7 @@
"openai/gpt-5.1-chat": {
"displayName": "GPT-5.1 Chat",
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
- "lifecycle": "active",
+ "lifecycle": "deprecated",
"contextWindow": 128000,
"maxOutputTokens": 64000,
"knowledgeCutoff": "2025-01-01",
@@ -37919,6 +39116,13 @@
"cacheReadUsdPer1M": 1,
"cacheWriteUsdPer1M": 12.5
},
+ {
+ "modelKey": "anthropic:claude-fable-5-1",
+ "inputUsdPer1M": 10,
+ "outputUsdPer1M": 50,
+ "cacheReadUsdPer1M": 0.25,
+ "cacheWriteUsdPer1M": 12.5
+ },
{
"modelKey": "anthropic:claude-haiku-4-5",
"inputUsdPer1M": 1,
@@ -38682,16 +39886,16 @@
"inputUsdPer1M": 0.058,
"outputUsdPer1M": 0.144
},
- {
- "modelKey": "cerebras:gemma-4-31b",
- "inputUsdPer1M": 0.99,
- "outputUsdPer1M": 1.49
- },
{
"modelKey": "cerebras:gpt-oss-120b",
"inputUsdPer1M": 0.35,
"outputUsdPer1M": 0.75
},
+ {
+ "modelKey": "cerebras:qwen-3.8-27b",
+ "inputUsdPer1M": 0.99,
+ "outputUsdPer1M": 1.49
+ },
{
"modelKey": "cohere:command-a-03-2025",
"inputUsdPer1M": 2.5,
@@ -38922,9 +40126,15 @@
},
{
"modelKey": "deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731",
- "inputUsdPer1M": 0.08,
+ "inputUsdPer1M": 0.06,
"outputUsdPer1M": 0.18,
- "cacheReadUsdPer1M": 0.016
+ "cacheReadUsdPer1M": 0.015
+ },
+ {
+ "modelKey": "deepinfra:deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
+ "inputUsdPer1M": 0.44,
+ "outputUsdPer1M": 1.32,
+ "cacheReadUsdPer1M": 0.014
},
{
"modelKey": "deepinfra:deepseek-ai/DeepSeek-V4-Pro",
@@ -38938,6 +40148,27 @@
"outputUsdPer1M": 2.6,
"cacheReadUsdPer1M": 0.1
},
+ {
+ "modelKey": "deepinfra:deepseek-ai/DeepSeek-V4.1-Flash",
+ "inputUsdPer1M": 0.2,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.006
+ },
+ {
+ "modelKey": "deepinfra:google/gemma-3-12b-it",
+ "inputUsdPer1M": 0.05,
+ "outputUsdPer1M": 0.15
+ },
+ {
+ "modelKey": "deepinfra:google/gemma-3-27b-it",
+ "inputUsdPer1M": 0.08,
+ "outputUsdPer1M": 0.16
+ },
+ {
+ "modelKey": "deepinfra:google/gemma-3-4b-it",
+ "inputUsdPer1M": 0.05,
+ "outputUsdPer1M": 0.1
+ },
{
"modelKey": "deepinfra:google/gemma-4-26B-A4B-it",
"inputUsdPer1M": 0.07,
@@ -39113,9 +40344,15 @@
},
{
"modelKey": "deepinfra:Qwen/Qwen3.8-27B",
- "inputUsdPer1M": 0.4,
- "outputUsdPer1M": 3,
- "cacheReadUsdPer1M": 0.04
+ "inputUsdPer1M": 0.2,
+ "outputUsdPer1M": 2.5,
+ "cacheReadUsdPer1M": 0.05
+ },
+ {
+ "modelKey": "deepinfra:Qwen/Qwen3.8-Flash",
+ "inputUsdPer1M": 0.113,
+ "outputUsdPer1M": 0.382,
+ "cacheReadUsdPer1M": 0.0141
},
{
"modelKey": "deepinfra:Qwen/Qwen3.8-Max",
@@ -39149,9 +40386,9 @@
},
{
"modelKey": "deepinfra:XiaomiMiMo/MiMo-V2.5",
- "inputUsdPer1M": 0.4,
- "outputUsdPer1M": 2,
- "cacheReadUsdPer1M": 0.08
+ "inputUsdPer1M": 0.14,
+ "outputUsdPer1M": 0.28,
+ "cacheReadUsdPer1M": 0.0028
},
{
"modelKey": "deepinfra:XiaomiMiMo/MiMo-V2.5-Pro",
@@ -39199,7 +40436,7 @@
"modelKey": "deepinfra:zai-org/GLM-5.3",
"inputUsdPer1M": 1.2,
"outputUsdPer1M": 4,
- "cacheReadUsdPer1M": 0.12
+ "cacheReadUsdPer1M": 0.2
},
{
"modelKey": "deepinfra:zai-org/GLM-5.3-Flash",
@@ -39207,17 +40444,23 @@
"outputUsdPer1M": 0.5,
"cacheReadUsdPer1M": 0.03
},
+ {
+ "modelKey": "deepseek:deepseek-flash",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.003
+ },
{
"modelKey": "deepseek:deepseek-v4-flash",
- "inputUsdPer1M": 0.14,
- "outputUsdPer1M": 0.28,
- "cacheReadUsdPer1M": 0.0028
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.003
},
{
"modelKey": "deepseek:deepseek-v4-flash-vision-exp",
- "inputUsdPer1M": 0.14,
- "outputUsdPer1M": 0.28,
- "cacheReadUsdPer1M": 0.0028
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.003
},
{
"modelKey": "deepseek:deepseek-v4-pro",
@@ -39227,9 +40470,15 @@
},
{
"modelKey": "fireworks-ai:accounts/fireworks/models/deepseek-v4-flash-0731",
- "inputUsdPer1M": 0.14,
- "outputUsdPer1M": 0.28,
- "cacheReadUsdPer1M": 0.028
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
+ },
+ {
+ "modelKey": "fireworks-ai:accounts/fireworks/models/deepseek-v4-flash-vision-exp",
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
},
{
"modelKey": "fireworks-ai:accounts/fireworks/models/deepseek-v4-pro-0813",
@@ -39237,6 +40486,12 @@
"outputUsdPer1M": 3.96,
"cacheReadUsdPer1M": 0.044
},
+ {
+ "modelKey": "fireworks-ai:accounts/fireworks/models/deepseek-v4p1-flash",
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
+ },
{
"modelKey": "fireworks-ai:accounts/fireworks/models/glm-5p2",
"inputUsdPer1M": 1.4,
@@ -39253,7 +40508,7 @@
"modelKey": "fireworks-ai:accounts/fireworks/models/glm-5p3-flash",
"inputUsdPer1M": 0.15,
"outputUsdPer1M": 0.5,
- "cacheReadUsdPer1M": 0.029
+ "cacheReadUsdPer1M": 0.03
},
{
"modelKey": "fireworks-ai:accounts/fireworks/models/gpt-oss-120b",
@@ -39315,6 +40570,12 @@
"outputUsdPer1M": 1.6,
"cacheReadUsdPer1M": 0.08
},
+ {
+ "modelKey": "fireworks-ai:accounts/fireworks/models/qwen3p8-2p4t-a95b",
+ "inputUsdPer1M": 2,
+ "outputUsdPer1M": 6,
+ "cacheReadUsdPer1M": 0.25
+ },
{
"modelKey": "fireworks-ai:accounts/fireworks/models/qwen3p8-max",
"inputUsdPer1M": 2,
@@ -39327,6 +40588,12 @@
"outputUsdPer1M": 6.6,
"cacheReadUsdPer1M": 0.21
},
+ {
+ "modelKey": "fireworks-ai:accounts/fireworks/routers/glm-5p3-fast",
+ "inputUsdPer1M": 2.1,
+ "outputUsdPer1M": 6.6,
+ "cacheReadUsdPer1M": 0.39
+ },
{
"modelKey": "fireworks-ai:accounts/fireworks/routers/kimi-k3-fast",
"inputUsdPer1M": 4.5,
@@ -39443,6 +40710,12 @@
"outputUsdPer1M": 3.75,
"cacheReadUsdPer1M": 0.075
},
+ {
+ "modelKey": "google:gemini-3.8-flash",
+ "inputUsdPer1M": 0.75,
+ "outputUsdPer1M": 3.75,
+ "cacheReadUsdPer1M": 0.075
+ },
{
"modelKey": "google:gemini-embedding-001",
"inputUsdPer1M": 0.15,
@@ -39573,6 +40846,11 @@
"inputUsdPer1M": 0.14,
"outputUsdPer1M": 0.28
},
+ {
+ "modelKey": "huggingface:deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
+ "inputUsdPer1M": 0.44,
+ "outputUsdPer1M": 1.32
+ },
{
"modelKey": "huggingface:deepseek-ai/DeepSeek-V4-Pro",
"inputUsdPer1M": 0.435,
@@ -39584,6 +40862,26 @@
"inputUsdPer1M": 1.32,
"outputUsdPer1M": 3.96
},
+ {
+ "modelKey": "huggingface:deepseek-ai/DeepSeek-V4.1-Flash",
+ "inputUsdPer1M": 0.3,
+ "outputUsdPer1M": 1.2
+ },
+ {
+ "modelKey": "huggingface:google/gemma-3-12b-it",
+ "inputUsdPer1M": 0.05,
+ "outputUsdPer1M": 0.15
+ },
+ {
+ "modelKey": "huggingface:google/gemma-3-27b-it",
+ "inputUsdPer1M": 0.08,
+ "outputUsdPer1M": 0.16
+ },
+ {
+ "modelKey": "huggingface:google/gemma-3-4b-it",
+ "inputUsdPer1M": 0.05,
+ "outputUsdPer1M": 0.1
+ },
{
"modelKey": "huggingface:google/gemma-4-26B-A4B-it",
"inputUsdPer1M": 0.13,
@@ -40104,61 +41402,49 @@
"cacheReadUsdPer1M": 0.14
},
{
- "modelKey": "moonshot:kimi-k2-0711-preview",
- "inputUsdPer1M": 0.6,
- "outputUsdPer1M": 2.5,
- "cacheReadUsdPer1M": 0.15
- },
- {
- "modelKey": "moonshot:kimi-k2-0905-preview",
- "inputUsdPer1M": 0.6,
- "outputUsdPer1M": 2.5,
- "cacheReadUsdPer1M": 0.15
+ "modelKey": "moonshot:kimi-k2.6",
+ "inputUsdPer1M": 0.95,
+ "outputUsdPer1M": 4,
+ "cacheReadUsdPer1M": 0.16
},
{
- "modelKey": "moonshot:kimi-k2-thinking",
- "inputUsdPer1M": 0.6,
- "outputUsdPer1M": 2.5,
- "cacheReadUsdPer1M": 0.15
+ "modelKey": "moonshot:kimi-k2.7-code",
+ "inputUsdPer1M": 0.95,
+ "outputUsdPer1M": 4,
+ "cacheReadUsdPer1M": 0.19
},
{
- "modelKey": "moonshot:kimi-k2-thinking-turbo",
- "inputUsdPer1M": 1.15,
+ "modelKey": "moonshot:kimi-k2.7-code-highspeed",
+ "inputUsdPer1M": 1.9,
"outputUsdPer1M": 8,
- "cacheReadUsdPer1M": 0.15
- },
- {
- "modelKey": "moonshot:kimi-k2-turbo-preview",
- "inputUsdPer1M": 2.4,
- "outputUsdPer1M": 10,
- "cacheReadUsdPer1M": 0.6
+ "cacheReadUsdPer1M": 0.38
},
{
- "modelKey": "moonshot:kimi-k2.5",
- "inputUsdPer1M": 0.6,
- "outputUsdPer1M": 3,
- "cacheReadUsdPer1M": 0.1
+ "modelKey": "moonshot:kimi-k3",
+ "inputUsdPer1M": 3,
+ "outputUsdPer1M": 15,
+ "cacheReadUsdPer1M": 0.3
},
{
- "modelKey": "moonshot:kimi-k2.6",
+ "modelKey": "moonshot-global:kimi-k2.6",
"inputUsdPer1M": 0.95,
"outputUsdPer1M": 4,
"cacheReadUsdPer1M": 0.16
},
{
- "modelKey": "moonshot:kimi-k2.7-code",
+ "modelKey": "moonshot-global:kimi-k2.7-code",
"inputUsdPer1M": 0.95,
"outputUsdPer1M": 4,
"cacheReadUsdPer1M": 0.19
},
{
- "modelKey": "moonshot:kimi-k2.7-code-highspeed",
+ "modelKey": "moonshot-global:kimi-k2.7-code-highspeed",
"inputUsdPer1M": 1.9,
"outputUsdPer1M": 8,
"cacheReadUsdPer1M": 0.38
},
{
- "modelKey": "moonshot:kimi-k3",
+ "modelKey": "moonshot-global:kimi-k3",
"inputUsdPer1M": 3,
"outputUsdPer1M": 15,
"cacheReadUsdPer1M": 0.3
@@ -40681,6 +41967,140 @@
"inputUsdPer1M": 0,
"outputUsdPer1M": 0
},
+ {
+ "modelKey": "nvidia:z-ai/glm-5.3-flash",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "ollama-cloud:deepseek-v4-flash",
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
+ },
+ {
+ "modelKey": "ollama-cloud:deepseek-v4-flash:0731",
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
+ },
+ {
+ "modelKey": "ollama-cloud:deepseek-v4-pro",
+ "inputUsdPer1M": 0.66,
+ "outputUsdPer1M": 1.98,
+ "cacheReadUsdPer1M": 0.022
+ },
+ {
+ "modelKey": "ollama-cloud:deepseek-v4-pro:0813",
+ "inputUsdPer1M": 0.66,
+ "outputUsdPer1M": 1.98,
+ "cacheReadUsdPer1M": 0.022
+ },
+ {
+ "modelKey": "ollama-cloud:deepseek-v4.1-flash",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.003
+ },
+ {
+ "modelKey": "ollama-cloud:gemma4:31b",
+ "inputUsdPer1M": 0.14,
+ "outputUsdPer1M": 0.4,
+ "cacheReadUsdPer1M": 0.05
+ },
+ {
+ "modelKey": "ollama-cloud:glm-5.1",
+ "inputUsdPer1M": 1,
+ "outputUsdPer1M": 3.2,
+ "cacheReadUsdPer1M": 0.2
+ },
+ {
+ "modelKey": "ollama-cloud:glm-5.2",
+ "inputUsdPer1M": 1.4,
+ "outputUsdPer1M": 4.4,
+ "cacheReadUsdPer1M": 0.26
+ },
+ {
+ "modelKey": "ollama-cloud:glm-5.3",
+ "inputUsdPer1M": 1.4,
+ "outputUsdPer1M": 4.4,
+ "cacheReadUsdPer1M": 0.26
+ },
+ {
+ "modelKey": "ollama-cloud:glm-5.3-flash",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.5,
+ "cacheReadUsdPer1M": 0.03
+ },
+ {
+ "modelKey": "ollama-cloud:gpt-oss:120b",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.014
+ },
+ {
+ "modelKey": "ollama-cloud:gpt-oss:20b",
+ "inputUsdPer1M": 0.07,
+ "outputUsdPer1M": 0.3,
+ "cacheReadUsdPer1M": 0.035
+ },
+ {
+ "modelKey": "ollama-cloud:kimi-k2.6",
+ "inputUsdPer1M": 0.95,
+ "outputUsdPer1M": 4,
+ "cacheReadUsdPer1M": 0.16
+ },
+ {
+ "modelKey": "ollama-cloud:kimi-k2.7-code",
+ "inputUsdPer1M": 0.95,
+ "outputUsdPer1M": 4,
+ "cacheReadUsdPer1M": 0.19
+ },
+ {
+ "modelKey": "ollama-cloud:kimi-k3",
+ "inputUsdPer1M": 3,
+ "outputUsdPer1M": 15,
+ "cacheReadUsdPer1M": 0.3
+ },
+ {
+ "modelKey": "ollama-cloud:minimax-m2.7",
+ "inputUsdPer1M": 0.3,
+ "outputUsdPer1M": 1.2,
+ "cacheReadUsdPer1M": 0.06
+ },
+ {
+ "modelKey": "ollama-cloud:minimax-m3",
+ "inputUsdPer1M": 0.6,
+ "outputUsdPer1M": 2.4,
+ "cacheReadUsdPer1M": 0.12
+ },
+ {
+ "modelKey": "ollama-cloud:mistral-large-3:675b",
+ "inputUsdPer1M": 0.5,
+ "outputUsdPer1M": 1.5
+ },
+ {
+ "modelKey": "ollama-cloud:nemotron-3-nano:30b",
+ "inputUsdPer1M": 0.06,
+ "outputUsdPer1M": 0.24
+ },
+ {
+ "modelKey": "ollama-cloud:nemotron-3-super",
+ "inputUsdPer1M": 0.015,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.015
+ },
+ {
+ "modelKey": "ollama-cloud:nemotron-3-ultra",
+ "inputUsdPer1M": 0.1,
+ "outputUsdPer1M": 3,
+ "cacheReadUsdPer1M": 0.1
+ },
+ {
+ "modelKey": "ollama-cloud:qwen3.5:397b",
+ "inputUsdPer1M": 0.6,
+ "outputUsdPer1M": 3.6
+ },
{
"modelKey": "openai:gpt-3.5-turbo",
"inputUsdPer1M": 0.5,
@@ -40902,6 +42322,13 @@
"cacheReadUsdPer1M": 1,
"cacheWriteUsdPer1M": 12.5
},
+ {
+ "modelKey": "opencode:claude-fable-5-1",
+ "inputUsdPer1M": 10,
+ "outputUsdPer1M": 50,
+ "cacheReadUsdPer1M": 0.25,
+ "cacheWriteUsdPer1M": 12.5
+ },
{
"modelKey": "opencode:claude-haiku-4-5",
"inputUsdPer1M": 1,
@@ -40977,6 +42404,12 @@
"outputUsdPer1M": 0,
"cacheReadUsdPer1M": 0
},
+ {
+ "modelKey": "opencode:deepseek-v4-flash-vision-exp",
+ "inputUsdPer1M": 0.14,
+ "outputUsdPer1M": 0.28,
+ "cacheReadUsdPer1M": 0.028
+ },
{
"modelKey": "opencode:deepseek-v4-pro",
"inputUsdPer1M": 1.74,
@@ -41013,6 +42446,12 @@
"outputUsdPer1M": 7.5,
"cacheReadUsdPer1M": 0.15
},
+ {
+ "modelKey": "opencode:gemini-3.8-flash",
+ "inputUsdPer1M": 1.5,
+ "outputUsdPer1M": 7.5,
+ "cacheReadUsdPer1M": 0.15
+ },
{
"modelKey": "opencode:glm-4.6",
"inputUsdPer1M": 0.6,
@@ -41055,6 +42494,18 @@
"outputUsdPer1M": 4.4,
"cacheReadUsdPer1M": 0.26
},
+ {
+ "modelKey": "opencode:glm-5.3",
+ "inputUsdPer1M": 1.4,
+ "outputUsdPer1M": 4.4,
+ "cacheReadUsdPer1M": 0.26
+ },
+ {
+ "modelKey": "opencode:glm-5.3-flash",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.5,
+ "cacheReadUsdPer1M": 0.03
+ },
{
"modelKey": "opencode:gpt-5",
"inputUsdPer1M": 1.07,
@@ -41324,6 +42775,18 @@
"outputUsdPer1M": 0,
"cacheReadUsdPer1M": 0
},
+ {
+ "modelKey": "opencode:muse-spark-1.3",
+ "inputUsdPer1M": 1.25,
+ "outputUsdPer1M": 4.25,
+ "cacheReadUsdPer1M": 0.15
+ },
+ {
+ "modelKey": "opencode:muse-spark-1.3-contributor-free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0,
+ "cacheReadUsdPer1M": 0
+ },
{
"modelKey": "opencode:nemotron-3-super-free",
"inputUsdPer1M": 0,
@@ -41392,7 +42855,7 @@
"modelKey": "openrouter:~anthropic/claude-fable-latest",
"inputUsdPer1M": 10,
"outputUsdPer1M": 50,
- "cacheReadUsdPer1M": 1,
+ "cacheReadUsdPer1M": 0.25,
"cacheWriteUsdPer1M": 12.5
},
{
@@ -41416,11 +42879,23 @@
"cacheReadUsdPer1M": 0.2,
"cacheWriteUsdPer1M": 2.5
},
+ {
+ "modelKey": "openrouter:~deepseek/deepseek-flash-latest",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.015
+ },
+ {
+ "modelKey": "openrouter:~deepseek/deepseek-pro-latest",
+ "inputUsdPer1M": 0.96,
+ "outputUsdPer1M": 2.88,
+ "cacheReadUsdPer1M": 0.088
+ },
{
"modelKey": "openrouter:~deepseek/deepseek-v4-flash-latest",
- "inputUsdPer1M": 0.05,
- "outputUsdPer1M": 0.16,
- "cacheReadUsdPer1M": 0.013
+ "inputUsdPer1M": 0.04,
+ "outputUsdPer1M": 0.1,
+ "cacheReadUsdPer1M": 0.01
},
{
"modelKey": "openrouter:~google/gemini-flash-latest",
@@ -41431,9 +42906,9 @@
},
{
"modelKey": "openrouter:~moonshotai/kimi-latest",
- "inputUsdPer1M": 2.55,
- "outputUsdPer1M": 12.75,
- "cacheReadUsdPer1M": 0.256
+ "inputUsdPer1M": 2.1,
+ "outputUsdPer1M": 10.95,
+ "cacheReadUsdPer1M": 0.23
},
{
"modelKey": "openrouter:~openai/gpt-mini-latest",
@@ -41441,11 +42916,17 @@
"outputUsdPer1M": 4.5,
"cacheReadUsdPer1M": 0.075
},
+ {
+ "modelKey": "openrouter:~z-ai/glm-flash-latest",
+ "inputUsdPer1M": 0.075,
+ "outputUsdPer1M": 0.25,
+ "cacheReadUsdPer1M": 0.015
+ },
{
"modelKey": "openrouter:~z-ai/glm-latest",
- "inputUsdPer1M": 1.17,
- "outputUsdPer1M": 3.96,
- "cacheReadUsdPer1M": 0.234
+ "inputUsdPer1M": 0.9,
+ "outputUsdPer1M": 3,
+ "cacheReadUsdPer1M": 0.15
},
{
"modelKey": "openrouter:aion-labs/aion-2.0",
@@ -41515,6 +42996,13 @@
"cacheReadUsdPer1M": 1,
"cacheWriteUsdPer1M": 12.5
},
+ {
+ "modelKey": "openrouter:anthropic/claude-fable-5.1",
+ "inputUsdPer1M": 10,
+ "outputUsdPer1M": 50,
+ "cacheReadUsdPer1M": 0.25,
+ "cacheWriteUsdPer1M": 12.5
+ },
{
"modelKey": "openrouter:anthropic/claude-haiku-4.5",
"inputUsdPer1M": 1,
@@ -41543,13 +43031,6 @@
"cacheReadUsdPer1M": 0.5,
"cacheWriteUsdPer1M": 6.25
},
- {
- "modelKey": "openrouter:anthropic/claude-opus-4.7-fast",
- "inputUsdPer1M": 30,
- "outputUsdPer1M": 150,
- "cacheReadUsdPer1M": 3,
- "cacheWriteUsdPer1M": 37.5
- },
{
"modelKey": "openrouter:anthropic/claude-opus-4.8",
"inputUsdPer1M": 5,
@@ -41557,13 +43038,6 @@
"cacheReadUsdPer1M": 0.5,
"cacheWriteUsdPer1M": 6.25
},
- {
- "modelKey": "openrouter:anthropic/claude-opus-4.8-fast",
- "inputUsdPer1M": 10,
- "outputUsdPer1M": 50,
- "cacheReadUsdPer1M": 1,
- "cacheWriteUsdPer1M": 12.5
- },
{
"modelKey": "openrouter:anthropic/claude-opus-5",
"inputUsdPer1M": 5,
@@ -41571,13 +43045,6 @@
"cacheReadUsdPer1M": 0.5,
"cacheWriteUsdPer1M": 6.25
},
- {
- "modelKey": "openrouter:anthropic/claude-opus-5-fast",
- "inputUsdPer1M": 10,
- "outputUsdPer1M": 50,
- "cacheReadUsdPer1M": 1,
- "cacheWriteUsdPer1M": 12.5
- },
{
"modelKey": "openrouter:anthropic/claude-sonnet-5",
"inputUsdPer1M": 2,
@@ -41649,9 +43116,9 @@
},
{
"modelKey": "openrouter:deepseek/deepseek-chat-v3.1",
- "inputUsdPer1M": 0.55,
- "outputUsdPer1M": 1.65,
- "cacheReadUsdPer1M": 0.55
+ "inputUsdPer1M": 0.25,
+ "outputUsdPer1M": 0.95,
+ "cacheReadUsdPer1M": 0.13
},
{
"modelKey": "openrouter:deepseek/deepseek-r1",
@@ -41688,21 +43155,21 @@
},
{
"modelKey": "openrouter:deepseek/deepseek-v4-flash",
- "inputUsdPer1M": 0.08092,
- "outputUsdPer1M": 0.16184,
- "cacheReadUsdPer1M": 0.016184
+ "inputUsdPer1M": 0.088606,
+ "outputUsdPer1M": 0.177212,
+ "cacheReadUsdPer1M": 0.017721
},
{
"modelKey": "openrouter:deepseek/deepseek-v4-flash-0731",
- "inputUsdPer1M": 0.065,
- "outputUsdPer1M": 0.18,
- "cacheReadUsdPer1M": 0.016
+ "inputUsdPer1M": 0.06,
+ "outputUsdPer1M": 0.12,
+ "cacheReadUsdPer1M": 0.012
},
{
"modelKey": "openrouter:deepseek/deepseek-v4-flash-vision-exp",
- "inputUsdPer1M": 0.44,
- "outputUsdPer1M": 1.32,
- "cacheReadUsdPer1M": 0.014
+ "inputUsdPer1M": 0.22,
+ "outputUsdPer1M": 0.66,
+ "cacheReadUsdPer1M": 0.007
},
{
"modelKey": "openrouter:deepseek/deepseek-v4-pro",
@@ -41712,9 +43179,15 @@
},
{
"modelKey": "openrouter:deepseek/deepseek-v4-pro-0813",
- "inputUsdPer1M": 1.1154,
- "outputUsdPer1M": 3.3462,
- "cacheReadUsdPer1M": 0.03718
+ "inputUsdPer1M": 0.9834,
+ "outputUsdPer1M": 2.9502,
+ "cacheReadUsdPer1M": 0.03278
+ },
+ {
+ "modelKey": "openrouter:deepseek/deepseek-v4.1-flash",
+ "inputUsdPer1M": 0.3,
+ "outputUsdPer1M": 1.2,
+ "cacheReadUsdPer1M": 0.006
},
{
"modelKey": "openrouter:dots-studio/dots-3-note-preview:free",
@@ -41820,6 +43293,13 @@
"cacheReadUsdPer1M": 0.075,
"cacheWriteUsdPer1M": 0.041667
},
+ {
+ "modelKey": "openrouter:google/gemini-3.8-flash",
+ "inputUsdPer1M": 0.75,
+ "outputUsdPer1M": 3.75,
+ "cacheReadUsdPer1M": 0.075,
+ "cacheWriteUsdPer1M": 0.041667
+ },
{
"modelKey": "openrouter:google/gemma-2-27b-it",
"inputUsdPer1M": 0.65,
@@ -41843,8 +43323,9 @@
},
{
"modelKey": "openrouter:google/gemma-4-26b-a4b-it",
- "inputUsdPer1M": 0.07,
- "outputUsdPer1M": 0.34
+ "inputUsdPer1M": 0.09,
+ "outputUsdPer1M": 0.3,
+ "cacheReadUsdPer1M": 0.05
},
{
"modelKey": "openrouter:google/gemma-4-26b-a4b-it:free",
@@ -41882,17 +43363,11 @@
"inputUsdPer1M": 0.017,
"outputUsdPer1M": 0.112
},
- {
- "modelKey": "openrouter:ibm-granite/granite-4.1-8b",
- "inputUsdPer1M": 0.05,
- "outputUsdPer1M": 0.1,
- "cacheReadUsdPer1M": 0.05
- },
{
"modelKey": "openrouter:ibm-granite/granite-4.2-8b",
- "inputUsdPer1M": 0.1,
- "outputUsdPer1M": 0.15,
- "cacheReadUsdPer1M": 0.05
+ "inputUsdPer1M": 0.06,
+ "outputUsdPer1M": 0.25,
+ "cacheReadUsdPer1M": 0.015
},
{
"modelKey": "openrouter:inception/mercury-2",
@@ -41900,17 +43375,57 @@
"outputUsdPer1M": 0.75,
"cacheReadUsdPer1M": 0.025
},
+ {
+ "modelKey": "openrouter:inception/mercury-2.5",
+ "inputUsdPer1M": 0.04,
+ "outputUsdPer1M": 0.15,
+ "cacheReadUsdPer1M": 0.004
+ },
{
"modelKey": "openrouter:inclusionai/ling-3.0-flash",
"inputUsdPer1M": 0.021,
"outputUsdPer1M": 0.063,
"cacheReadUsdPer1M": 0.0042
},
+ {
+ "modelKey": "openrouter:inclusionai/ling-3.0-flash-fin",
+ "inputUsdPer1M": 0.06,
+ "outputUsdPer1M": 0.18,
+ "cacheReadUsdPer1M": 0.012
+ },
{
"modelKey": "openrouter:inclusionai/ling-3.0-flash-fin:free",
"inputUsdPer1M": 0,
"outputUsdPer1M": 0
},
+ {
+ "modelKey": "openrouter:inclusionai/ling-3.0-flash-sante:free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "openrouter:inclusionai/ling-3.0-flash-vl",
+ "inputUsdPer1M": 0.06,
+ "outputUsdPer1M": 0.18,
+ "cacheReadUsdPer1M": 0.012
+ },
+ {
+ "modelKey": "openrouter:inclusionai/ling-3.0-flash-vl:free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "openrouter:inference-net/schematron-v2-small",
+ "inputUsdPer1M": 0.05,
+ "outputUsdPer1M": 0.23,
+ "cacheReadUsdPer1M": 0.05
+ },
+ {
+ "modelKey": "openrouter:inference-net/schematron-v2-turbo",
+ "inputUsdPer1M": 0.03,
+ "outputUsdPer1M": 0.15,
+ "cacheReadUsdPer1M": 0.03
+ },
{
"modelKey": "openrouter:kwaipilot/kat-coder-pro-v2",
"inputUsdPer1M": 0.3,
@@ -41962,14 +43477,13 @@
},
{
"modelKey": "openrouter:meta-llama/llama-3.3-70b-instruct",
- "inputUsdPer1M": 0.71,
- "outputUsdPer1M": 0.71,
- "cacheReadUsdPer1M": 0.71
+ "inputUsdPer1M": 0.1,
+ "outputUsdPer1M": 0.32
},
{
"modelKey": "openrouter:meta-llama/llama-4-maverick",
- "inputUsdPer1M": 0.2,
- "outputUsdPer1M": 0.696
+ "inputUsdPer1M": 0.1875,
+ "outputUsdPer1M": 0.6525
},
{
"modelKey": "openrouter:meta-llama/llama-4-scout",
@@ -41983,8 +43497,8 @@
},
{
"modelKey": "openrouter:meta/muse-glimmer-30b",
- "inputUsdPer1M": 0.3,
- "outputUsdPer1M": 1.2,
+ "inputUsdPer1M": 0.35,
+ "outputUsdPer1M": 1.5,
"cacheReadUsdPer1M": 0.04
},
{
@@ -42005,6 +43519,18 @@
"outputUsdPer1M": 0.2,
"cacheReadUsdPer1M": 0.002
},
+ {
+ "modelKey": "openrouter:meta/muse-spark-1.3",
+ "inputUsdPer1M": 1.25,
+ "outputUsdPer1M": 4.25,
+ "cacheReadUsdPer1M": 0.15
+ },
+ {
+ "modelKey": "openrouter:meta/muse-spark-1.3-contributor",
+ "inputUsdPer1M": 0.1,
+ "outputUsdPer1M": 0.2,
+ "cacheReadUsdPer1M": 0.002
+ },
{
"modelKey": "openrouter:microsoft/phi-4",
"inputUsdPer1M": 0.07,
@@ -42054,22 +43580,12 @@
"outputUsdPer1M": 1.2,
"cacheReadUsdPer1M": 0.06
},
- {
- "modelKey": "openrouter:minimax/minimax-m2.7:free",
- "inputUsdPer1M": 0,
- "outputUsdPer1M": 0
- },
{
"modelKey": "openrouter:minimax/minimax-m3",
"inputUsdPer1M": 0.3,
"outputUsdPer1M": 1.2,
"cacheReadUsdPer1M": 0.06
},
- {
- "modelKey": "openrouter:minimax/minimax-m3:free",
- "inputUsdPer1M": 0,
- "outputUsdPer1M": 0
- },
{
"modelKey": "openrouter:mistralai/codestral-2508",
"inputUsdPer1M": 0.3,
@@ -42209,15 +43725,15 @@
},
{
"modelKey": "openrouter:moonshotai/kimi-k2.7-code",
- "inputUsdPer1M": 0.66,
- "outputUsdPer1M": 3.4,
- "cacheReadUsdPer1M": 0.18
+ "inputUsdPer1M": 0.71,
+ "outputUsdPer1M": 3.5,
+ "cacheReadUsdPer1M": 0.15
},
{
"modelKey": "openrouter:moonshotai/kimi-k3",
- "inputUsdPer1M": 3,
- "outputUsdPer1M": 15,
- "cacheReadUsdPer1M": 0.3
+ "inputUsdPer1M": 2.648138,
+ "outputUsdPer1M": 13.282724,
+ "cacheReadUsdPer1M": 0.302644
},
{
"modelKey": "openrouter:morph/morph-v3-fast",
@@ -42230,16 +43746,14 @@
"outputUsdPer1M": 1.9
},
{
- "modelKey": "openrouter:nex-agi/nex-n2-mini",
- "inputUsdPer1M": 0.025,
- "outputUsdPer1M": 0.1,
- "cacheReadUsdPer1M": 0.0025
+ "modelKey": "openrouter:nex-agi/nex-n2.5-mini:free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
},
{
- "modelKey": "openrouter:nex-agi/nex-n2-pro",
- "inputUsdPer1M": 0.25,
- "outputUsdPer1M": 1,
- "cacheReadUsdPer1M": 0.025
+ "modelKey": "openrouter:nex-agi/nex-n2.5-pro:free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
},
{
"modelKey": "openrouter:nousresearch/hermes-3-llama-3.1-405b",
@@ -42256,16 +43770,11 @@
"inputUsdPer1M": 1,
"outputUsdPer1M": 3
},
- {
- "modelKey": "openrouter:nousresearch/hermes-4-70b",
- "inputUsdPer1M": 0.13,
- "outputUsdPer1M": 0.4
- },
{
"modelKey": "openrouter:nvidia/nemotron-3-nano-30b-a3b",
"inputUsdPer1M": 0.05,
"outputUsdPer1M": 0.2,
- "cacheReadUsdPer1M": 0.025
+ "cacheReadUsdPer1M": 0.03
},
{
"modelKey": "openrouter:nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
@@ -42274,8 +43783,8 @@
},
{
"modelKey": "openrouter:nvidia/nemotron-3-super-120b-a12b",
- "inputUsdPer1M": 0.085,
- "outputUsdPer1M": 0.4
+ "inputUsdPer1M": 0.08,
+ "outputUsdPer1M": 0.45
},
{
"modelKey": "openrouter:nvidia/nemotron-3-super-120b-a12b:free",
@@ -42284,15 +43793,20 @@
},
{
"modelKey": "openrouter:nvidia/nemotron-3-ultra-550b-a55b",
- "inputUsdPer1M": 0.5,
- "outputUsdPer1M": 2.2,
- "cacheReadUsdPer1M": 0.1
+ "inputUsdPer1M": 0.6,
+ "outputUsdPer1M": 2.4,
+ "cacheReadUsdPer1M": 0.12
},
{
"modelKey": "openrouter:nvidia/nemotron-3-ultra-550b-a55b:free",
"inputUsdPer1M": 0,
"outputUsdPer1M": 0
},
+ {
+ "modelKey": "openrouter:nvidia/nemotron-3.5-content-safety",
+ "inputUsdPer1M": 0.2,
+ "outputUsdPer1M": 0.2
+ },
{
"modelKey": "openrouter:nvidia/nemotron-3.5-content-safety:free",
"inputUsdPer1M": 0,
@@ -42339,11 +43853,6 @@
"inputUsdPer1M": 10,
"outputUsdPer1M": 30
},
- {
- "modelKey": "openrouter:openai/gpt-4-turbo-preview",
- "inputUsdPer1M": 10,
- "outputUsdPer1M": 30
- },
{
"modelKey": "openrouter:openai/gpt-4.1",
"inputUsdPer1M": 2,
@@ -42656,8 +44165,9 @@
},
{
"modelKey": "openrouter:qwen/qwen2.5-vl-72b-instruct",
- "inputUsdPer1M": 0.25,
- "outputUsdPer1M": 0.75
+ "inputUsdPer1M": 0.8,
+ "outputUsdPer1M": 1,
+ "cacheReadUsdPer1M": 0.4
},
{
"modelKey": "openrouter:qwen/qwen3-14b",
@@ -42724,9 +44234,8 @@
},
{
"modelKey": "openrouter:qwen/qwen3-next-80b-a3b-instruct",
- "inputUsdPer1M": 0.1,
- "outputUsdPer1M": 1.1,
- "cacheReadUsdPer1M": 0.07
+ "inputUsdPer1M": 0.09,
+ "outputUsdPer1M": 1.1
},
{
"modelKey": "openrouter:qwen/qwen3-next-80b-a3b-thinking",
@@ -42771,8 +44280,8 @@
},
{
"modelKey": "openrouter:qwen/qwen3.5-122b-a10b",
- "inputUsdPer1M": 0.29,
- "outputUsdPer1M": 2.4
+ "inputUsdPer1M": 0.26,
+ "outputUsdPer1M": 2.08
},
{
"modelKey": "openrouter:qwen/qwen3.5-27b",
@@ -42781,14 +44290,14 @@
},
{
"modelKey": "openrouter:qwen/qwen3.5-35b-a3b",
- "inputUsdPer1M": 0.25,
- "outputUsdPer1M": 1.25,
- "cacheReadUsdPer1M": 0.25
+ "inputUsdPer1M": 0.1625,
+ "outputUsdPer1M": 1.3
},
{
"modelKey": "openrouter:qwen/qwen3.5-397b-a17b",
- "inputUsdPer1M": 0.39,
- "outputUsdPer1M": 2.34
+ "inputUsdPer1M": 0.55,
+ "outputUsdPer1M": 3.5,
+ "cacheReadUsdPer1M": 0.225
},
{
"modelKey": "openrouter:qwen/qwen3.5-9b",
@@ -42802,9 +44311,9 @@
},
{
"modelKey": "openrouter:qwen/qwen3.6-27b",
- "inputUsdPer1M": 0.6,
- "outputUsdPer1M": 3.6,
- "cacheReadUsdPer1M": 0.12
+ "inputUsdPer1M": 0.3,
+ "outputUsdPer1M": 2,
+ "cacheReadUsdPer1M": 0.03
},
{
"modelKey": "openrouter:qwen/qwen3.6-35b-a3b",
@@ -42827,10 +44336,9 @@
},
{
"modelKey": "openrouter:qwen/qwen3.8-27b",
- "inputUsdPer1M": 0.425,
+ "inputUsdPer1M": 0.214,
"outputUsdPer1M": 2.55,
- "cacheReadUsdPer1M": 0.085,
- "cacheWriteUsdPer1M": 0.53125
+ "cacheReadUsdPer1M": 0.15
},
{
"modelKey": "openrouter:qwen/qwen3.8-flash",
@@ -42840,7 +44348,7 @@
"cacheWriteUsdPer1M": 0.2
},
{
- "modelKey": "openrouter:qwen/qwen3.8-max",
+ "modelKey": "openrouter:qwen/qwen3.8-max-0902",
"inputUsdPer1M": 2,
"outputUsdPer1M": 6,
"cacheReadUsdPer1M": 0.25,
@@ -42866,6 +44374,12 @@
"inputUsdPer1M": 1,
"outputUsdPer1M": 3
},
+ {
+ "modelKey": "openrouter:sakana/fugu-max",
+ "inputUsdPer1M": 2,
+ "outputUsdPer1M": 6,
+ "cacheReadUsdPer1M": 0.25
+ },
{
"modelKey": "openrouter:sakana/sakana-namazu",
"inputUsdPer1M": 0.95,
@@ -42977,7 +44491,7 @@
},
{
"modelKey": "openrouter:undi95/remm-slerp-l2-13b",
- "inputUsdPer1M": 0.45,
+ "inputUsdPer1M": 0.35,
"outputUsdPer1M": 0.65
},
{
@@ -42988,9 +44502,9 @@
},
{
"modelKey": "openrouter:upstage/solar-pro4",
- "inputUsdPer1M": 0.03,
- "outputUsdPer1M": 0.12,
- "cacheReadUsdPer1M": 0.006
+ "inputUsdPer1M": 0.09,
+ "outputUsdPer1M": 0.36,
+ "cacheReadUsdPer1M": 0.018
},
{
"modelKey": "openrouter:writer/palmyra-x5",
@@ -43047,9 +44561,8 @@
},
{
"modelKey": "openrouter:z-ai/glm-4.7-flash",
- "inputUsdPer1M": 0.06,
- "outputUsdPer1M": 0.4,
- "cacheReadUsdPer1M": 0.01
+ "inputUsdPer1M": 0.0605,
+ "outputUsdPer1M": 0.4
},
{
"modelKey": "openrouter:z-ai/glm-5",
@@ -43071,14 +44584,9 @@
},
{
"modelKey": "openrouter:z-ai/glm-5.2",
- "inputUsdPer1M": 1.19,
- "outputUsdPer1M": 3.74,
- "cacheReadUsdPer1M": 0.221
- },
- {
- "modelKey": "openrouter:z-ai/glm-5.2:free",
- "inputUsdPer1M": 0,
- "outputUsdPer1M": 0
+ "inputUsdPer1M": 1.4,
+ "outputUsdPer1M": 4.4,
+ "cacheReadUsdPer1M": 0.14
},
{
"modelKey": "openrouter:z-ai/glm-5.3",
@@ -43088,9 +44596,9 @@
},
{
"modelKey": "openrouter:z-ai/glm-5.3-flash",
- "inputUsdPer1M": 0.075,
- "outputUsdPer1M": 0.25,
- "cacheReadUsdPer1M": 0.015
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.5,
+ "cacheReadUsdPer1M": 0.03
},
{
"modelKey": "openrouter:z-ai/glm-5v-turbo",
@@ -43454,6 +44962,12 @@
"outputUsdPer1M": 3.96,
"cacheReadUsdPer1M": 0.13
},
+ {
+ "modelKey": "togetherai:deepseek-ai/DeepSeek-V4.1-Flash",
+ "inputUsdPer1M": 0.3,
+ "outputUsdPer1M": 1.2,
+ "cacheReadUsdPer1M": 0.006
+ },
{
"modelKey": "togetherai:essentialai/Rnj-1-Instruct",
"inputUsdPer1M": 0.15,
@@ -43667,7 +45181,7 @@
"modelKey": "vercel:alibaba/qwen-3.6-max-preview",
"inputUsdPer1M": 1.3,
"outputUsdPer1M": 7.8,
- "cacheReadUsdPer1M": 0.26,
+ "cacheReadUsdPer1M": 0.13,
"cacheWriteUsdPer1M": 1.625
},
{
@@ -43744,13 +45258,13 @@
"modelKey": "vercel:alibaba/qwen3.5-flash",
"inputUsdPer1M": 0.1,
"outputUsdPer1M": 0.4,
- "cacheReadUsdPer1M": 0.001,
+ "cacheReadUsdPer1M": 0.01,
"cacheWriteUsdPer1M": 0.125
},
{
"modelKey": "vercel:alibaba/qwen3.5-plus",
"inputUsdPer1M": 0.4,
- "outputUsdPer1M": 2.4,
+ "outputUsdPer1M": 2.5,
"cacheReadUsdPer1M": 0.04,
"cacheWriteUsdPer1M": 0.5
},
@@ -43763,7 +45277,7 @@
"modelKey": "vercel:alibaba/qwen3.6-plus",
"inputUsdPer1M": 0.5,
"outputUsdPer1M": 3,
- "cacheReadUsdPer1M": 0.1,
+ "cacheReadUsdPer1M": 0.05,
"cacheWriteUsdPer1M": 0.625
},
{
@@ -43802,19 +45316,20 @@
},
{
"modelKey": "vercel:alibaba/qwen3.8-flash",
- "inputUsdPer1M": 0.16,
+ "inputUsdPer1M": 0.15,
"outputUsdPer1M": 0.47,
"cacheReadUsdPer1M": 0.016,
"cacheWriteUsdPer1M": 0.2
},
{
- "modelKey": "vercel:alibaba/qwen3.8-flash-next",
- "inputUsdPer1M": 0.12,
- "outputUsdPer1M": 0.4,
- "cacheReadUsdPer1M": 0.01
+ "modelKey": "vercel:alibaba/qwen3.8-max",
+ "inputUsdPer1M": 2,
+ "outputUsdPer1M": 6,
+ "cacheReadUsdPer1M": 0.25,
+ "cacheWriteUsdPer1M": 2.5
},
{
- "modelKey": "vercel:alibaba/qwen3.8-max",
+ "modelKey": "vercel:alibaba/qwen3.8-max-0902",
"inputUsdPer1M": 2,
"outputUsdPer1M": 6,
"cacheReadUsdPer1M": 0.25,
@@ -43830,19 +45345,22 @@
"modelKey": "vercel:amazon/nova-lite",
"inputUsdPer1M": 0.06,
"outputUsdPer1M": 0.24,
- "cacheReadUsdPer1M": 0.015
+ "cacheReadUsdPer1M": 0.015,
+ "cacheWriteUsdPer1M": 0.06
},
{
"modelKey": "vercel:amazon/nova-micro",
"inputUsdPer1M": 0.035,
"outputUsdPer1M": 0.14,
- "cacheReadUsdPer1M": 0.00875
+ "cacheReadUsdPer1M": 0.00875,
+ "cacheWriteUsdPer1M": 0.035
},
{
"modelKey": "vercel:amazon/nova-pro",
"inputUsdPer1M": 0.8,
"outputUsdPer1M": 3.2,
- "cacheReadUsdPer1M": 0.2
+ "cacheReadUsdPer1M": 0.2,
+ "cacheWriteUsdPer1M": 0.8
},
{
"modelKey": "vercel:anthropic/claude-3-haiku",
@@ -43858,6 +45376,13 @@
"cacheReadUsdPer1M": 1,
"cacheWriteUsdPer1M": 12.5
},
+ {
+ "modelKey": "vercel:anthropic/claude-fable-5.1",
+ "inputUsdPer1M": 10,
+ "outputUsdPer1M": 50,
+ "cacheReadUsdPer1M": 0.25,
+ "cacheWriteUsdPer1M": 12.5
+ },
{
"modelKey": "vercel:anthropic/claude-haiku-4.5",
"inputUsdPer1M": 1,
@@ -43959,6 +45484,12 @@
"outputUsdPer1M": 2,
"cacheReadUsdPer1M": 0.05
},
+ {
+ "modelKey": "vercel:bytedance/seed-2.1-turbo",
+ "inputUsdPer1M": 0.5,
+ "outputUsdPer1M": 2.5,
+ "cacheReadUsdPer1M": 0.1
+ },
{
"modelKey": "vercel:cohere/command-a",
"inputUsdPer1M": 2.5,
@@ -43969,12 +45500,6 @@
"inputUsdPer1M": 1.35,
"outputUsdPer1M": 5.4
},
- {
- "modelKey": "vercel:deepseek/deepseek-v3",
- "inputUsdPer1M": 0.27,
- "outputUsdPer1M": 1.12,
- "cacheReadUsdPer1M": 0.135
- },
{
"modelKey": "vercel:deepseek/deepseek-v3.1",
"inputUsdPer1M": 0.25,
@@ -43989,9 +45514,8 @@
},
{
"modelKey": "vercel:deepseek/deepseek-v3.2",
- "inputUsdPer1M": 0.28,
- "outputUsdPer1M": 0.42,
- "cacheReadUsdPer1M": 0.028
+ "inputUsdPer1M": 0.62,
+ "outputUsdPer1M": 1.85
},
{
"modelKey": "vercel:deepseek/deepseek-v3.2-thinking",
@@ -44028,6 +45552,12 @@
"outputUsdPer1M": 1.98,
"cacheReadUsdPer1M": 0.066
},
+ {
+ "modelKey": "vercel:deepseek/deepseek-v4.1-flash",
+ "inputUsdPer1M": 0.15,
+ "outputUsdPer1M": 0.6,
+ "cacheReadUsdPer1M": 0.015
+ },
{
"modelKey": "vercel:google/gemini-2.5-flash",
"inputUsdPer1M": 0.3,
@@ -44117,6 +45647,12 @@
"outputUsdPer1M": 3.75,
"cacheReadUsdPer1M": 0.075
},
+ {
+ "modelKey": "vercel:google/gemini-3.8-flash",
+ "inputUsdPer1M": 0.75,
+ "outputUsdPer1M": 3.75,
+ "cacheReadUsdPer1M": 0.075
+ },
{
"modelKey": "vercel:google/gemini-omni-flash-preview",
"inputUsdPer1M": 1.5,
@@ -44139,6 +45675,12 @@
"outputUsdPer1M": 0.75,
"cacheReadUsdPer1M": 0.024999999999999998
},
+ {
+ "modelKey": "vercel:inception/mercury-2.5",
+ "inputUsdPer1M": 0.04,
+ "outputUsdPer1M": 0.15,
+ "cacheReadUsdPer1M": 0.004
+ },
{
"modelKey": "vercel:inception/mercury-coder-small",
"inputUsdPer1M": 0.25,
@@ -44146,9 +45688,9 @@
},
{
"modelKey": "vercel:inclusionai/ling-3.0-flash",
- "inputUsdPer1M": 0.06,
- "outputUsdPer1M": 0.18,
- "cacheReadUsdPer1M": 0.012
+ "inputUsdPer1M": 0.021,
+ "outputUsdPer1M": 0.063,
+ "cacheReadUsdPer1M": 0.0042
},
{
"modelKey": "vercel:inclusionai/ling-3.0-flash-fin",
@@ -44160,6 +45702,26 @@
"inputUsdPer1M": 0,
"outputUsdPer1M": 0
},
+ {
+ "modelKey": "vercel:inclusionai/ling-3.0-flash-sante",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "vercel:inclusionai/ling-3.0-flash-sante-free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "vercel:inclusionai/ling-3.0-flash-vl",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
+ {
+ "modelKey": "vercel:inclusionai/ling-3.0-flash-vl-free",
+ "inputUsdPer1M": 0,
+ "outputUsdPer1M": 0
+ },
{
"modelKey": "vercel:interfaze/interfaze-beta",
"inputUsdPer1M": 1.5,
@@ -44238,6 +45800,18 @@
"outputUsdPer1M": 0.2,
"cacheReadUsdPer1M": 0.002
},
+ {
+ "modelKey": "vercel:meta/muse-spark-1.3",
+ "inputUsdPer1M": 1.25,
+ "outputUsdPer1M": 4.25,
+ "cacheReadUsdPer1M": 0.15
+ },
+ {
+ "modelKey": "vercel:meta/muse-spark-1.3-contributor",
+ "inputUsdPer1M": 0.1,
+ "outputUsdPer1M": 0.2,
+ "cacheReadUsdPer1M": 0.002
+ },
{
"modelKey": "vercel:minimax/minimax-m2",
"inputUsdPer1M": 0.3,
@@ -44280,11 +45854,6 @@
"cacheReadUsdPer1M": 0.06,
"cacheWriteUsdPer1M": 0.375
},
- {
- "modelKey": "vercel:minimax/minimax-m2.7-free",
- "inputUsdPer1M": 0,
- "outputUsdPer1M": 0
- },
{
"modelKey": "vercel:minimax/minimax-m2.7-highspeed",
"inputUsdPer1M": 0.6,
@@ -44298,12 +45867,6 @@
"outputUsdPer1M": 1.2,
"cacheReadUsdPer1M": 0.06
},
- {
- "modelKey": "vercel:minimax/minimax-m3-free",
- "inputUsdPer1M": 0,
- "outputUsdPer1M": 0,
- "cacheReadUsdPer1M": 0
- },
{
"modelKey": "vercel:mistral/codestral",
"inputUsdPer1M": 0.3,
@@ -44424,7 +45987,8 @@
{
"modelKey": "vercel:nvidia/nemotron-3-nano-30b-a3b",
"inputUsdPer1M": 0.05,
- "outputUsdPer1M": 0.24
+ "outputUsdPer1M": 0.2,
+ "cacheReadUsdPer1M": 0.025
},
{
"modelKey": "vercel:nvidia/nemotron-3-super-120b-a12b",
@@ -44703,7 +46267,7 @@
"inputUsdPer1M": 0.4,
"outputUsdPer1M": 2.4,
"cacheReadUsdPer1M": 0.04,
- "cacheWriteUsdPer1M": 0.25
+ "cacheWriteUsdPer1M": 0.5
},
{
"modelKey": "vercel:openai/gpt-5.6-sol",
@@ -44717,7 +46281,7 @@
"inputUsdPer1M": 4,
"outputUsdPer1M": 20,
"cacheReadUsdPer1M": 0.4,
- "cacheWriteUsdPer1M": 2.5
+ "cacheWriteUsdPer1M": 5
},
{
"modelKey": "vercel:openai/gpt-5.6-terra",
@@ -44731,7 +46295,7 @@
"inputUsdPer1M": 4,
"outputUsdPer1M": 24,
"cacheReadUsdPer1M": 0.4,
- "cacheWriteUsdPer1M": 2.5
+ "cacheWriteUsdPer1M": 5
},
{
"modelKey": "vercel:openai/gpt-image-1",
@@ -44757,6 +46321,18 @@
"outputUsdPer1M": 30,
"cacheReadUsdPer1M": 1.25
},
+ {
+ "modelKey": "vercel:openai/gpt-image-2.5-flare",
+ "inputUsdPer1M": 5,
+ "outputUsdPer1M": 30,
+ "cacheReadUsdPer1M": 1.25
+ },
+ {
+ "modelKey": "vercel:openai/gpt-image-2.5-sunburst",
+ "inputUsdPer1M": 5,
+ "outputUsdPer1M": 30,
+ "cacheReadUsdPer1M": 1.25
+ },
{
"modelKey": "vercel:openai/gpt-oss-120b",
"inputUsdPer1M": 0.1,
@@ -44853,12 +46429,24 @@
"inputUsdPer1M": 0,
"outputUsdPer1M": 0
},
+ {
+ "modelKey": "vercel:sakana/fugu-max",
+ "inputUsdPer1M": 2,
+ "outputUsdPer1M": 6,
+ "cacheReadUsdPer1M": 0.25
+ },
{
"modelKey": "vercel:sakana/fugu-ultra",
"inputUsdPer1M": 5,
"outputUsdPer1M": 30,
"cacheReadUsdPer1M": 0.5
},
+ {
+ "modelKey": "vercel:sakana/fugu-ultra-v2",
+ "inputUsdPer1M": 5,
+ "outputUsdPer1M": 30,
+ "cacheReadUsdPer1M": 0.5
+ },
{
"modelKey": "vercel:sakana/namazu",
"inputUsdPer1M": 0.95,
@@ -45000,12 +46588,6 @@
"outputUsdPer1M": 0.87,
"cacheReadUsdPer1M": 0.0036
},
- {
- "modelKey": "vercel:xiaomi/mimo-v2.5-pro-ultraspeed",
- "inputUsdPer1M": 1.305,
- "outputUsdPer1M": 2.61,
- "cacheReadUsdPer1M": 0.0108
- },
{
"modelKey": "vercel:zai/glm-4.5",
"inputUsdPer1M": 0.6,
@@ -45082,6 +46664,12 @@
"outputUsdPer1M": 4.4,
"cacheReadUsdPer1M": 0.14
},
+ {
+ "modelKey": "vercel:zai/glm-5.3-fast",
+ "inputUsdPer1M": 2.1,
+ "outputUsdPer1M": 6.6,
+ "cacheReadUsdPer1M": 0.21
+ },
{
"modelKey": "vercel:zai/glm-5.3-flash",
"inputUsdPer1M": 0.15,
@@ -45990,7 +47578,7 @@
"id": "kimi-for-coding",
"name": "Kimi For Coding",
"api": "https://api.kimi.com/coding/v1",
- "doc": "https://www.kimi.com/code/docs/en/third-party-tools/other-coding-agents.html"
+ "doc": "https://www.kimi.com/code/docs/en/kimi-code/models.html"
},
"MiniMax": {
"id": "minimax",
@@ -46021,6 +47609,12 @@
"api": "https://api.moonshot.cn/v1",
"doc": "https://platform.moonshot.cn/docs/api/chat"
},
+ "moonshot-global": {
+ "id": "moonshotai",
+ "name": "Moonshot AI",
+ "api": "https://api.moonshot.ai/v1",
+ "doc": "https://platform.moonshot.ai/docs/api/chat"
+ },
"nvidia": {
"id": "nvidia",
"name": "Nvidia",
@@ -46191,6 +47785,7 @@
"minimax-coding-plan": {},
"mistral": {},
"moonshot": {},
+ "moonshot-global": {},
"nvidia": {},
"ollama-cloud": {},
"openai": {},
@@ -46201,6 +47796,9 @@
"claude-fable-5": {
"npm": "@ai-sdk/anthropic"
},
+ "claude-fable-5-1": {
+ "npm": "@ai-sdk/anthropic"
+ },
"claude-haiku-4-5": {
"npm": "@ai-sdk/anthropic"
},
@@ -46255,6 +47853,9 @@
"gemini-3.7-flash": {
"npm": "@ai-sdk/google"
},
+ "gemini-3.8-flash": {
+ "npm": "@ai-sdk/google"
+ },
"gpt-5": {
"npm": "@ai-sdk/openai"
},
@@ -46315,6 +47916,9 @@
"gpt-5.6-terra": {
"npm": "@ai-sdk/openai"
},
+ "gpt-6-astra": {
+ "npm": "@ai-sdk/openai"
+ },
"grok-4.5": {
"npm": "@ai-sdk/openai"
},
@@ -46339,6 +47943,12 @@
"muse-spark-1.2-contributor-free": {
"npm": "@ai-sdk/openai"
},
+ "muse-spark-1.3": {
+ "npm": "@ai-sdk/openai"
+ },
+ "muse-spark-1.3-contributor-free": {
+ "npm": "@ai-sdk/openai"
+ },
"qwen3.5-plus": {
"npm": "@ai-sdk/anthropic"
},
@@ -46371,6 +47981,9 @@
"muse-spark-1.2-contributor": {
"npm": "@ai-sdk/openai"
},
+ "muse-spark-1.3-contributor": {
+ "npm": "@ai-sdk/openai"
+ },
"qwen3.8-flash": {
"npm": "@ai-sdk/anthropic"
}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@maka.apache.org
For additional commands, e-mail: commits-help@maka.apache.org