Get started

How good, how fast, what it costs to pick wrong.

Almost every board on this page orders models by a figure someone else measured and published: quality by Artificial Analysis and Design Arena, speed by OpenRouter timing the traffic it routes, and seller spread by what the companies selling a model ask for it, which ranks nothing about the model, only how much the choice of seller is worth on it. The usage board is the exception and says so: the tokens each model ran through OpenRouter, as OpenRouter publishes them, which is adoption on one router rather than a measure of the model.

58 models scored by Artificial Analysis · 132 ranked by Design Arena · 324 timed by OpenRouter · 186 sold by more than one company · 84 ranked by OpenRouter usage · 359 models in the catalog·retrieved sha256 · 0bb53a77713f

  1. 01IntelligenceClaude Fable 5.264.6
  2. 03CodingClaude Fable 5.180.7
  3. 04AgenticClaude Fable 5.157.2
  4. 05Throughputgpt-oss-safeguard-20b438 tok/s
  5. 05First tokenGranite 4.2 8B148ms
  6. 06ValueLing 3.1 Flash439 per $
  7. 07Design ArenaMuse Wonder1378 Elo
  8. 12UsageDeepSeek V4.1 Flash 101723.9%
01 / 13

Claude Fable 5.2 leads at 64.6; 58 models scored.

025507510064.6Claude Fable 5.260.0Gemini 4 Carbon59.1Claude Opus 5.556.9GPT-6 Astra54.6GPT-6.1 Sol54.5Claude Sonnet 5.553.9Gemini 4 Hydrogen51.3GPT-6.1 Terra50.4Gemini 4 Argon50.0Claude Fable 5.146.5Kimi K3.1 Pro44.2Claude Opus 544.2GPT-6 Sol43.6Muse Wonder36.6Muse Spark 1.336.5Claude Haiku 5.535.5Grok 4.734.6Hy433.7Qwen3.8 Max (0902)32.3GPT-5.6 Terra31.7ZGLM 5.331.0Grok 4.630.8Gemini 3.8 Flash30.6TInkling 2 Small29.9MiMo-V2.6-Pro29.6Ling 3.1 FlashIntelligence Index, 0 to 100Claude Fable 5.2: 64.6 on the Intelligence Index · AnthropicClaude Fable 5.264.6 on the Intelligence Index · AnthropicGemini 4 Carbon: 60.0 on the Intelligence Index · GoogleGemini 4 Carbon60.0 on the Intelligence Index · GoogleClaude Opus 5.5: 59.1 on the Intelligence Index · AnthropicClaude Opus 5.559.1 on the Intelligence Index · AnthropicGPT-6 Astra: 56.9 on the Intelligence Index · OpenAIGPT-6 Astra56.9 on the Intelligence Index · OpenAIGPT-6.1 Sol: 54.6 on the Intelligence Index · OpenAIGPT-6.1 Sol54.6 on the Intelligence Index · OpenAIClaude Sonnet 5.5: 54.5 on the Intelligence Index · AnthropicClaude Sonnet 5.554.5 on the Intelligence Index · AnthropicGemini 4 Hydrogen: 53.9 on the Intelligence Index · GoogleGemini 4 Hydrogen53.9 on the Intelligence Index · GoogleGPT-6.1 Terra: 51.3 on the Intelligence Index · OpenAIGPT-6.1 Terra51.3 on the Intelligence Index · OpenAIGemini 4 Argon: 50.4 on the Intelligence Index · GoogleGemini 4 Argon50.4 on the Intelligence Index · GoogleClaude Fable 5.1: 50.0 on the Intelligence Index · AnthropicClaude Fable 5.150.0 on the Intelligence Index · AnthropicKimi K3.1 Pro: 46.5 on the Intelligence Index · MoonshotAIKimi K3.1 Pro46.5 on the Intelligence Index · MoonshotAIClaude Opus 5: 44.2 on the Intelligence Index · AnthropicClaude Opus 544.2 on the Intelligence Index · AnthropicGPT-6 Sol: 44.2 on the Intelligence Index · OpenAIGPT-6 Sol44.2 on the Intelligence Index · OpenAIMuse Wonder: 43.6 on the Intelligence Index · MetaMuse Wonder43.6 on the Intelligence Index · MetaMuse Spark 1.3: 36.6 on the Intelligence Index · MetaMuse Spark 1.336.6 on the Intelligence Index · MetaClaude Haiku 5.5: 36.5 on the Intelligence Index · AnthropicClaude Haiku 5.536.5 on the Intelligence Index · AnthropicGrok 4.7: 35.5 on the Intelligence Index · SpaceXAIGrok 4.735.5 on the Intelligence Index · SpaceXAIHy4: 34.6 on the Intelligence Index · TencentHy434.6 on the Intelligence Index · TencentQwen3.8 Max (0902): 33.7 on the Intelligence Index · QwenQwen3.8 Max (0902)33.7 on the Intelligence Index · QwenGPT-5.6 Terra: 32.3 on the Intelligence Index · OpenAIGPT-5.6 Terra32.3 on the Intelligence Index · OpenAIGLM 5.3: 31.7 on the Intelligence Index · Z.aiGLM 5.331.7 on the Intelligence Index · Z.aiGrok 4.6: 31.0 on the Intelligence Index · SpaceXAIGrok 4.631.0 on the Intelligence Index · SpaceXAIGemini 3.8 Flash: 30.8 on the Intelligence Index · GoogleGemini 3.8 Flash30.8 on the Intelligence Index · GoogleInkling 2 Small: 30.6 on the Intelligence Index · Thinking MachinesInkling 2 Small30.6 on the Intelligence Index · Thinking MachinesMiMo-V2.6-Pro: 29.9 on the Intelligence Index · XiaomiMiMo-V2.6-Pro29.9 on the Intelligence Index · XiaomiLing 3.1 Flash: 29.6 on the Intelligence Index · inclusionAILing 3.1 Flash29.6 on the Intelligence Index · inclusionAI
  1. 01Claude Fable 5.2Anthropic64.6
  2. 02Gemini 4 CarbonGoogle60.0
  3. 03Claude Opus 5.5Anthropic59.1
  4. 04GPT-6 AstraOpenAI56.9
  5. 05GPT-6.1 SolOpenAI54.6
  6. 06Claude Sonnet 5.5Anthropic54.5
  7. 07Gemini 4 HydrogenGoogle53.9
  8. 08GPT-6.1 TerraOpenAI51.3
  9. 09Gemini 4 ArgonGoogle50.4
  10. 10Claude Fable 5.1Anthropic50.0
  11. 11Kimi K3.1 ProMoonshotAI46.5
  12. 12Claude Opus 5Anthropic44.2
  13. 13GPT-6 SolOpenAI44.2
  14. 14Muse WonderMeta43.6
  15. 15Muse Spark 1.3Meta36.6
  16. 16Claude Haiku 5.5Anthropic36.5
  17. 17Grok 4.7SpaceXAI35.5
  18. 18Hy4Tencent34.6
  19. 19Qwen3.8 Max (0902)Qwen33.7
  20. 20GPT-5.6 TerraOpenAI32.3
  21. 21GLM 5.3Z.ai31.7
  22. 22Grok 4.6SpaceXAI31.0
  23. 23Gemini 3.8 FlashGoogle30.8
  24. 24Inkling 2 SmallThinking Machines30.6
  25. 25MiMo-V2.6-ProXiaomi29.9
  26. 26Ling 3.1 FlashinclusionAI29.6

The top 26 of 58, 64.6 down to 29.6 on the Intelligence Index. Every column stands on zero on the index's own 0 to 100 scale, so a column half as tall is half the score, and the maker's mark hangs under each. 32 more are scored and ranked on the intelligence leaderboard.

02 / 13

13 models have held the Agentic Index ceiling since Mar 2025; Claude Fable 5.1 holds it at 57.2.

  • a scored model, in its maker's tone
  • raised the ceiling, named
  • the running best
Jul 2025Jan 2026Jul 20260204060Listed on OpenRouter, Mar 2025 to 17 Nov 2026Agentic IndexGemma 3 12B: 0.1 · listed 13 Mar 2025 · Google13 Mar 20250.1Gemma 3 12B0.1 · listed 13 Mar 2025 · GoogleLlama 4 Maverick: 0.6 · listed 5 Apr 2025 · Meta Llama5 Apr 20250.6Llama 4 Maverick0.6 · listed 5 Apr 2025 · Meta LlamaLlama 4 Scout: 0.5 · listed 5 Apr 2025 · Meta Llama5 Apr 20250.5Llama 4 Scout0.5 · listed 5 Apr 2025 · Meta LlamaQwen3 32B: 0.9 · listed 28 Apr 2025 · Qwen28 Apr 20250.9Qwen3 32B0.9 · listed 28 Apr 2025 · QwenQwen3 235B A22B Thinking 2507: 1.3 · listed 25 Jul 2025 · Qwen25 Jul 20251.3Qwen3 235B A22B Thinking 25071.3 · listed 25 Jul 2025 · Qwengpt-oss-20b: 1.2 · listed 5 Aug 2025 · OpenAI5 Aug 20251.2gpt-oss-20b1.2 · listed 5 Aug 2025 · OpenAIMistral Medium 3.1: 1.9 · listed 13 Aug 2025 · Mistral13 Aug 20251.9Mistral Medium 3.11.9 · listed 13 Aug 2025 · MistralDeepSeek V3.1 Terminus: 6.8 · listed 22 Sep 2025 · DeepSeek22 Sep 20256.8DeepSeek V3.1 Terminus6.8 · listed 22 Sep 2025 · DeepSeekClaude Haiku 4.5: 8.0 · listed 15 Oct 2025 · Anthropic15 Oct 20258.0Claude Haiku 4.58.0 · listed 15 Oct 2025 · AnthropicMistral Large 3 2512: 1.0 · listed 1 Dec 2025 · Mistral1 Dec 20251.0Mistral Large 3 25121.0 · listed 1 Dec 2025 · MistralMinistral 3 14B 2512: 1.1 · listed 2 Dec 2025 · Mistral2 Dec 20251.1Ministral 3 14B 25121.1 · listed 2 Dec 2025 · MistralMinistral 3 3B 2512: 0.8 · listed 2 Dec 2025 · Mistral2 Dec 20250.8Ministral 3 3B 25120.8 · listed 2 Dec 2025 · MistralMinistral 3 8B 2512: 0.6 · listed 2 Dec 2025 · Mistral2 Dec 20250.6Ministral 3 8B 25120.6 · listed 2 Dec 2025 · MistralDevstral 2 2512: 2.3 · listed 9 Dec 2025 · Mistral9 Dec 20252.3Devstral 2 25122.3 · listed 9 Dec 2025 · MistralNemotron 3 Nano 30B A3B: 1.0 · listed 14 Dec 2025 · NVIDIA14 Dec 20251.0Nemotron 3 Nano 30B A3B1.0 · listed 14 Dec 2025 · NVIDIAQwen3 Coder Next: 0.9 · listed 4 Feb 2026 · Qwen4 Feb 20260.9Qwen3 Coder Next0.9 · listed 4 Feb 2026 · QwenQwen3.5 397B A17B: 8.3 · listed 16 Feb 2026 · Qwen16 Feb 20268.3Qwen3.5 397B A17B8.3 · listed 16 Feb 2026 · QwenGemini 3.1 Pro Preview: 8.2 · listed 19 Feb 2026 · Google19 Feb 20268.2Gemini 3.1 Pro Preview8.2 · listed 19 Feb 2026 · GoogleQwen3.5-122B-A10B: 7.6 · listed 25 Feb 2026 · Qwen25 Feb 20267.6Qwen3.5-122B-A10B7.6 · listed 25 Feb 2026 · QwenGemini 3.1 Flash Lite Preview: 1.6 · listed 3 Mar 2026 · Google3 Mar 20261.6Gemini 3.1 Flash Lite Preview1.6 · listed 3 Mar 2026 · GoogleQwen3.5-9B: 1.2 · listed 10 Mar 2026 · Qwen10 Mar 20261.2Qwen3.5-9B1.2 · listed 10 Mar 2026 · QwenNemotron 3 Super: 1.7 · listed 11 Mar 2026 · NVIDIA11 Mar 20261.7Nemotron 3 Super1.7 · listed 11 Mar 2026 · NVIDIAMistral Small 4: 0.8 · listed 16 Mar 2026 · Mistral16 Mar 20260.8Mistral Small 40.8 · listed 16 Mar 2026 · MistralGPT-5.4 Mini: 17.9 · listed 17 Mar 2026 · OpenAI17 Mar 202617.9GPT-5.4 Mini17.9 · listed 17 Mar 2026 · OpenAIGPT-5.4 Nano: 16.0 · listed 17 Mar 2026 · OpenAI17 Mar 202616.0GPT-5.4 Nano16.0 · listed 17 Mar 2026 · OpenAIMiniMax M2.7: 15.3 · listed 18 Mar 2026 · MiniMax18 Mar 202615.3MiniMax M2.715.3 · listed 18 Mar 2026 · MiniMaxGemma 4 31B: 4.2 · listed 2 Apr 2026 · Google2 Apr 20264.2Gemma 4 31B4.2 · listed 2 Apr 2026 · GoogleGemma 4 26B A4B: 2.9 · listed 3 Apr 2026 · Google3 Apr 20262.9Gemma 4 26B A4B2.9 · listed 3 Apr 2026 · GoogleGLM 5.1: 23.9 · listed 7 Apr 2026 · Z.ai7 Apr 202623.9GLM 5.123.9 · listed 7 Apr 2026 · Z.aiKimi K2.6: 20.5 · listed 20 Apr 2026 · MoonshotAI20 Apr 202620.5Kimi K2.620.5 · listed 20 Apr 2026 · MoonshotAIHy3 preview: 24.1 · listed 22 Apr 2026 · Tencent22 Apr 202624.1Hy3 preview24.1 · listed 22 Apr 2026 · TencentMiMo-V2.5: 15.8 · listed 22 Apr 2026 · Xiaomi22 Apr 202615.8MiMo-V2.515.8 · listed 22 Apr 2026 · XiaomiMiMo-V2.5-Pro: 21.3 · listed 22 Apr 2026 · Xiaomi22 Apr 202621.3MiMo-V2.5-Pro21.3 · listed 22 Apr 2026 · XiaomiDeepSeek V4 Flash 0423: 22.2 · listed 24 Apr 2026 · DeepSeek24 Apr 202622.2DeepSeek V4 Flash 042322.2 · listed 24 Apr 2026 · DeepSeekGPT-5.5: 36.4 · listed 24 Apr 2026 · OpenAI24 Apr 202636.4GPT-5.536.4 · listed 24 Apr 2026 · OpenAIQwen3.6 27B: 18.5 · listed 27 Apr 2026 · Qwen27 Apr 202618.5Qwen3.6 27B18.5 · listed 27 Apr 2026 · QwenQwen3.6 35B A3B: 13.1 · listed 27 Apr 2026 · Qwen27 Apr 202613.1Qwen3.6 35B A3B13.1 · listed 27 Apr 2026 · QwenMistral Medium 3.5: 7.0 · listed 30 Apr 2026 · Mistral30 Apr 20267.0Mistral Medium 3.57.0 · listed 30 Apr 2026 · MistralGrok 4.3: 15.5 · listed 30 Apr 2026 · SpaceXAI30 Apr 202615.5Grok 4.315.5 · listed 30 Apr 2026 · SpaceXAIGemini 3.5 Flash: 26.0 · listed 19 May 2026 · Google19 May 202626.0Gemini 3.5 Flash26.0 · listed 19 May 2026 · GoogleQwen3.7 Max: 22.5 · listed 21 May 2026 · Qwen21 May 202622.5Qwen3.7 Max22.5 · listed 21 May 2026 · QwenMiniMax M3: 29.5 · listed 31 May 2026 · MiniMax31 May 202629.5MiniMax M329.5 · listed 31 May 2026 · MiniMaxQwen3.7 Plus: 17.5 · listed 3 Jun 2026 · Qwen3 Jun 202617.5Qwen3.7 Plus17.5 · listed 3 Jun 2026 · QwenNemotron 3 Ultra: 20.1 · listed 4 Jun 2026 · NVIDIA4 Jun 202620.1Nemotron 3 Ultra20.1 · listed 4 Jun 2026 · NVIDIAKimi K2.7 Code: 21.0 · listed 12 Jun 2026 · MoonshotAI12 Jun 202621.0Kimi K2.7 Code21.0 · listed 12 Jun 2026 · MoonshotAIGLM 5.2: 38.4 · listed 16 Jun 2026 · Z.ai16 Jun 202638.4GLM 5.238.4 · listed 16 Jun 2026 · Z.aiNorth Mini Code (free): 1.1 · listed 17 Jun 2026 · Cohere17 Jun 20261.1North Mini Code (free)1.1 · listed 17 Jun 2026 · CohereClaude Sonnet 5: 43.6 · listed 30 Jun 2026 · Anthropic30 Jun 202643.6Claude Sonnet 543.6 · listed 30 Jun 2026 · AnthropicGrok 4.5: 41.2 · listed 8 Jul 2026 · SpaceXAI8 Jul 202641.2Grok 4.541.2 · listed 8 Jul 2026 · SpaceXAIGPT-5.6 Luna: 42.1 · listed 9 Jul 2026 · OpenAI9 Jul 202642.1GPT-5.6 Luna42.1 · listed 9 Jul 2026 · OpenAIGPT-5.6 Sol: 47.4 · listed 9 Jul 2026 · OpenAI9 Jul 202647.4GPT-5.6 Sol47.4 · listed 9 Jul 2026 · OpenAIGPT-5.6 Terra: 43.2 · listed 9 Jul 2026 · OpenAI9 Jul 202643.2GPT-5.6 Terra43.2 · listed 9 Jul 2026 · OpenAIMuse Spark 1.1: 25.8 · listed 16 Jul 2026 · Meta16 Jul 202625.8Muse Spark 1.125.8 · listed 16 Jul 2026 · MetaKimi K3: 29.3 · listed 16 Jul 2026 · MoonshotAI16 Jul 202629.3Kimi K329.3 · listed 16 Jul 2026 · MoonshotAIInkling: 22.5 · listed 17 Jul 2026 · Thinking Machines17 Jul 202622.5Inkling22.5 · listed 17 Jul 2026 · Thinking MachinesLongCat 2.0: 14.0 · listed 20 Jul 2026 · Meituan20 Jul 202614.0LongCat 2.014.0 · listed 20 Jul 2026 · MeituanGemini 3.5 Flash Lite: 14.3 · listed 21 Jul 2026 · Google21 Jul 202614.3Gemini 3.5 Flash Lite14.3 · listed 21 Jul 2026 · GoogleGemini 3.6 Flash: 29.0 · listed 21 Jul 2026 · Google21 Jul 202629.0Gemini 3.6 Flash29.0 · listed 21 Jul 2026 · GoogleLing 3.0 Flash: 19.3 · listed 23 Jul 2026 · inclusionAI23 Jul 202619.3Ling 3.0 Flash19.3 · listed 23 Jul 2026 · inclusionAIInkling Small: 23.5 · listed 30 Jul 2026 · Thinking Machines30 Jul 202623.5Inkling Small23.5 · listed 30 Jul 2026 · Thinking MachinesDeepSeek V4 Flash 0731: 41.0 · listed 31 Jul 2026 · DeepSeek31 Jul 202641.0DeepSeek V4 Flash 073141.0 · listed 31 Jul 2026 · DeepSeekMuse Spark 1.2: 43.2 · listed 5 Aug 2026 · Meta5 Aug 202643.2Muse Spark 1.243.2 · listed 5 Aug 2026 · MetaLing 3.0 Tiny: 4.4 · listed 6 Aug 2026 · inclusionAI6 Aug 20264.4Ling 3.0 Tiny4.4 · listed 6 Aug 2026 · inclusionAIMuse Glimmer 30B: 8.5 · listed 9 Aug 2026 · Meta9 Aug 20268.5Muse Glimmer 30B8.5 · listed 9 Aug 2026 · MetaNemotron 3.5 Lightning: 3.5 · listed 11 Aug 2026 · NVIDIA11 Aug 20263.5Nemotron 3.5 Lightning3.5 · listed 11 Aug 2026 · NVIDIADeepSeek V4 Pro 0813: 41.3 · listed 12 Aug 2026 · DeepSeek12 Aug 202641.3DeepSeek V4 Pro 081341.3 · listed 12 Aug 2026 · DeepSeekQwen3.8 2.4T A95B: 50.1 · listed 12 Aug 2026 · Qwen12 Aug 202650.1Qwen3.8 2.4T A95B50.1 · listed 12 Aug 2026 · QwenGrok 4.6: 52.5 · listed 12 Aug 2026 · SpaceXAI12 Aug 202652.5Grok 4.652.5 · listed 12 Aug 2026 · SpaceXAIGemini 3.7 Flash: 35.2 · listed 13 Aug 2026 · Google13 Aug 202635.2Gemini 3.7 Flash35.2 · listed 13 Aug 2026 · GoogleQwen3.8 27B: 45.8 · listed 14 Aug 2026 · Qwen14 Aug 202645.8Qwen3.8 27B45.8 · listed 14 Aug 2026 · QwenGLM 5.3: 53.1 · listed 18 Aug 2026 · Z.ai18 Aug 202653.1GLM 5.353.1 · listed 18 Aug 2026 · Z.aiDeepSeek V4 Flash Vision Exp: 47.5 · listed 21 Aug 2026 · DeepSeek21 Aug 202647.5DeepSeek V4 Flash Vision Exp47.5 · listed 21 Aug 2026 · DeepSeekGLM 5.3 Flash: 50.9 · listed 26 Aug 2026 · Z.ai26 Aug 202650.9GLM 5.3 Flash50.9 · listed 26 Aug 2026 · Z.aiGranite 4.2 8B: 1.3 · listed 31 Aug 2026 · IBM Granite31 Aug 20261.3Granite 4.2 8B1.3 · listed 31 Aug 2026 · IBM GraniteGemini 3.8 Flash: 40.2 · listed 2 Sep 2026 · Google2 Sep 202640.2Gemini 3.8 Flash40.2 · listed 2 Sep 2026 · GoogleMuse Spark 1.3: 55.5 · listed 2 Sep 2026 · Meta2 Sep 202655.5Muse Spark 1.355.5 · listed 2 Sep 2026 · MetaQwen3.8 Max (0902): 56.0 · listed 3 Sep 2026 · Qwen3 Sep 202656.0Qwen3.8 Max (0902)56.0 · listed 3 Sep 2026 · QwenGPT-6 Astra: 51.0 · listed 4 Sep 2026 · OpenAI4 Sep 202651.0GPT-6 Astra51.0 · listed 4 Sep 2026 · OpenAILing 3.0 Flash VL: 28.7 · listed 10 Sep 2026 · inclusionAI10 Sep 202628.7Ling 3.0 Flash VL28.7 · listed 10 Sep 2026 · inclusionAICommand A+: 1.0 · listed 22 Sep 2026 · Cohere22 Sep 20261.0Command A+1.0 · listed 22 Sep 2026 · CohereMotif 3: 28.4 · listed 2 Nov 2026 · Motif Technologies2 Nov 202628.4Motif 328.4 · listed 2 Nov 2026 · Motif TechnologiesGemma 3 27B: 0.1 · listed 12 Mar 2025 · Google12 Mar 20250.101 Gemma 3 27B0.1 · listed 12 Mar 2025 · GoogleDeepSeek V3 0324: 0.8 · listed 24 Mar 2025 · DeepSeek24 Mar 20250.802 DeepSeek V3 03240.8 · listed 24 Mar 2025 · DeepSeekQwen3 14B: 0.9 · listed 28 Apr 2025 · Qwen28 Apr 20250.903 Qwen3 14B0.9 · listed 28 Apr 2025 · QwenGemini 2.5 Pro: 1.6 · listed 17 Jun 2025 · Google17 Jun 20251.604 Gemini 2.5 Pro1.6 · listed 17 Jun 2025 · Googlegpt-oss-120b: 3.7 · listed 5 Aug 2025 · OpenAI5 Aug 20253.705 gpt-oss-120b3.7 · listed 5 Aug 2025 · OpenAIGPT-5 Mini: 6.8 · listed 7 Aug 2025 · OpenAI7 Aug 20256.806 GPT-5 Mini6.8 · listed 7 Aug 2025 · OpenAIClaude Sonnet 4.5: 15.8 · listed 29 Sep 2025 · Anthropic29 Sep 202515.807 Claude Sonnet 4.515.8 · listed 29 Sep 2025 · AnthropicClaude Sonnet 4.6: 31.8 · listed 17 Feb 2026 · Anthropic17 Feb 202631.808 Claude Sonnet 4.631.8 · listed 17 Feb 2026 · AnthropicClaude Opus 4.7: 38.6 · listed 16 Apr 2026 · Anthropic16 Apr 202638.609 Claude Opus 4.738.6 · listed 16 Apr 2026 · AnthropicClaude Opus 4.8: 41.9 · listed 27 May 2026 · Anthropic27 May 202641.910 Claude Opus 4.841.9 · listed 27 May 2026 · AnthropicClaude Fable 5: 50.7 · listed 9 Jun 2026 · Anthropic9 Jun 202650.711 Claude Fable 550.7 · listed 9 Jun 2026 · AnthropicClaude Opus 5: 56.5 · listed 24 Jul 2026 · Anthropic24 Jul 202656.512 Claude Opus 556.5 · listed 24 Jul 2026 · AnthropicClaude Fable 5.1: 57.2 · listed 1 Sep 2026 · Anthropic1 Sep 202657.213 Claude Fable 5.157.2 · listed 1 Sep 2026 · Anthropic
  1. 01Gemma 3 27B0.1 · listed 12 Mar 2025
  2. 02DeepSeek V3 03240.8 · listed 24 Mar 2025
  3. 03Qwen3 14B0.9 · listed 28 Apr 2025
  4. 04Gemini 2.5 Pro1.6 · listed 17 Jun 2025
  5. 05gpt-oss-120b3.7 · listed 5 Aug 2025
  6. 06GPT-5 Mini6.8 · listed 7 Aug 2025
  7. 07Claude Sonnet 4.515.8 · listed 29 Sep 2025
  8. 08Claude Sonnet 4.631.8 · listed 17 Feb 2026
  9. 09Claude Opus 4.738.6 · listed 16 Apr 2026
  10. 10Claude Opus 4.841.9 · listed 27 May 2026
  11. 11Claude Fable 550.7 · listed 9 Jun 2026
  12. 12Claude Opus 556.5 · listed 24 Jul 2026
  13. 13Claude Fable 5.157.2 · listed 1 Sep 2026

Each dot is one of the 94 scored models, at the date OpenRouter listed it, which is not always the date the provider released it, filled in its maker's tone as the columns above are. The staircase is the running best, and the models that raised it are named on the figure and numbered in the key. The axis runs from zero to 60, the first round tick over the ceiling, rather than to the index's 100, because nothing here is a length. Scores are the Agentic Index as measured today; the catalog does not carry past scores, so this lays the current measurement over the listing dates rather than replaying history.

03 / 13

Claude Fable 5.1, 80.7.

  1. 01Claude Fable 5.1Anthropic80.7
  2. 02GPT-5.6 SolOpenAI78.3
  3. 03Claude Opus 5Anthropic78.0
  4. 04GPT-6 AstraOpenAI76.9
  5. 05GPT-5.6 TerraOpenAI76.7
  6. 06Claude Fable 5Anthropic76.5
  7. 07Gemini 3.8 FlashGoogle76.3
  8. 08Qwen3.8 Max (0902)Qwen76.2
  9. 09Gemini 3.7 FlashGoogle76.1
  10. 10Grok 4.6SpaceXAI75.9
  11. 11Muse Spark 1.3Meta75.8
  12. 12GPT-5.5OpenAI74.9
  13. 13GLM 5.3Z.ai74.8
  14. 14Claude Opus 4.8Anthropic74.3
  15. 15Claude Opus 4.7Anthropic73.6
  16. 16Grok 4.5SpaceXAI72.4
  17. 17Muse Spark 1.2Meta72.2
  18. 18Kimi K3MoonshotAI72.0
  19. 19Qwen3.8 2.4T A95BQwen71.9
  20. 20Claude Sonnet 5Anthropic71.5
all 118 scored
04 / 13

Claude Fable 5.1, 57.2.

  1. 01Claude Fable 5.1Anthropic57.2
  2. 02Claude Opus 5Anthropic56.5
  3. 03Qwen3.8 Max (0902)Qwen56.0
  4. 04Muse Spark 1.3Meta55.5
  5. 05GLM 5.3Z.ai53.1
  6. 06Grok 4.6SpaceXAI52.5
  7. 07GPT-6 AstraOpenAI51.0
  8. 08GLM 5.3 FlashZ.ai50.9
  9. 09Claude Fable 5Anthropic50.7
  10. 10Qwen3.8 2.4T A95BQwen50.1
  11. 11DeepSeek V4 Flash Vision ExpDeepSeek47.5
  12. 12GPT-5.6 SolOpenAI47.4
  13. 13Qwen3.8 27BQwen45.8
  14. 14Claude Sonnet 5Anthropic43.6
  15. 15Muse Spark 1.2Meta43.2
  16. 16GPT-5.6 TerraOpenAI43.2
  17. 17GPT-5.6 LunaOpenAI42.1
  18. 18Claude Opus 4.8Anthropic41.9
  19. 19DeepSeek V4 Pro 0813DeepSeek41.3
  20. 20Grok 4.5SpaceXAI41.2
all 94 scored
05 / 13

gpt-oss-safeguard-20b writes fastest at 438 tok/s; Granite 4.2 8B answers first at 148ms.

0200400438gpt-oss-safeguard-20b432LFM2.5-24B-A2B417RBeam 1.0 501B A23B290LFM3-1.7B266Llama 3.1 8B Instruct260Nemotron 3.5 Lightning249gpt-oss-20b231Nemotron 3 Nano 30B A3B224MMiniMax M2.7224Muse Spark Open223gpt-oss-120b223Muse Ember214Nemotron 3.5 Super VL204Nemotron 3 Super192TInkling 2 Small186Gemini 2.5 Flash Lite172Gemini 3.1 Flash Lite Preview167Ministral 3 3B 2512150DeepSeek V4.1 Flash148Nova Micro 1.0Output tokens per second, mediangpt-oss-safeguard-20b: 438 tok/s over 8,747 requests · OpenAIgpt-oss-safeguard-20b438 tok/s over 8,747 requests · OpenAILFM2.5-24B-A2B: 432 tok/s over 377 requests · LiquidAILFM2.5-24B-A2B432 tok/s over 377 requests · LiquidAIBeam 1.0 501B A23B: 417 tok/s over 845 requests · ReflectionBeam 1.0 501B A23B417 tok/s over 845 requests · ReflectionLFM3-1.7B: 290 tok/s over 3,747 requests · LiquidAILFM3-1.7B290 tok/s over 3,747 requests · LiquidAILlama 3.1 8B Instruct: 266 tok/s over 75,207 requests · Meta LlamaLlama 3.1 8B Instruct266 tok/s over 75,207 requests · Meta LlamaNemotron 3.5 Lightning: 260 tok/s over 10,681 requests · NVIDIANemotron 3.5 Lightning260 tok/s over 10,681 requests · NVIDIAgpt-oss-20b: 249 tok/s over 93,041 requests · OpenAIgpt-oss-20b249 tok/s over 93,041 requests · OpenAINemotron 3 Nano 30B A3B: 231 tok/s over 1,259 requests · NVIDIANemotron 3 Nano 30B A3B231 tok/s over 1,259 requests · NVIDIAMiniMax M2.7: 224 tok/s over 2,146 requests · MiniMaxMiniMax M2.7224 tok/s over 2,146 requests · MiniMaxMuse Spark Open: 224 tok/s over 1,551 requests · MetaMuse Spark Open224 tok/s over 1,551 requests · Metagpt-oss-120b: 223 tok/s over 225,515 requests · OpenAIgpt-oss-120b223 tok/s over 225,515 requests · OpenAIMuse Ember: 223 tok/s over 55,876 requests · MetaMuse Ember223 tok/s over 55,876 requests · MetaNemotron 3.5 Super VL: 214 tok/s over 13,683 requests · NVIDIANemotron 3.5 Super VL214 tok/s over 13,683 requests · NVIDIANemotron 3 Super: 204 tok/s over 998 requests · NVIDIANemotron 3 Super204 tok/s over 998 requests · NVIDIAInkling 2 Small: 192 tok/s over 14,293 requests · Thinking MachinesInkling 2 Small192 tok/s over 14,293 requests · Thinking MachinesGemini 2.5 Flash Lite: 186 tok/s over 274,468 requests · GoogleGemini 2.5 Flash Lite186 tok/s over 274,468 requests · GoogleGemini 3.1 Flash Lite Preview: 172 tok/s over 31,653 requests · GoogleGemini 3.1 Flash Lite Preview172 tok/s over 31,653 requests · GoogleMinistral 3 3B 2512: 167 tok/s over 47,942 requests · MistralMinistral 3 3B 2512167 tok/s over 47,942 requests · MistralDeepSeek V4.1 Flash: 150 tok/s over 1,389,152 requests · DeepSeekDeepSeek V4.1 Flash150 tok/s over 1,389,152 requests · DeepSeekNova Micro 1.0: 148 tok/s over 13,967 requests · AmazonNova Micro 1.0148 tok/s over 13,967 requests · Amazon
Throughput

Output tokens per second, median.

  1. 01gpt-oss-safeguard-20bOpenAI438
  2. 02LFM2.5-24B-A2BLiquidAI432
  3. 03Beam 1.0 501B A23BReflection417
  4. 04LFM3-1.7BLiquidAI290
  5. 05Llama 3.1 8B InstructMeta Llama266
  6. 06Nemotron 3.5 LightningNVIDIA260
  7. 07gpt-oss-20bOpenAI249
  8. 08Nemotron 3 Nano 30B A3BNVIDIA231
  9. 09MiniMax M2.7MiniMax224
  10. 10Muse Spark OpenMeta224
  11. 11gpt-oss-120bOpenAI223
  12. 12Muse EmberMeta223
  13. 13Nemotron 3.5 Super VLNVIDIA214
  14. 14Nemotron 3 SuperNVIDIA204
  15. 15Inkling 2 SmallThinking Machines192
  16. 16Gemini 2.5 Flash LiteGoogle186
  17. 17Gemini 3.1 Flash Lite PreviewGoogle172
  18. 18Ministral 3 3B 2512Mistral167
  19. 19DeepSeek V4.1 FlashDeepSeek150
  20. 20Nova Micro 1.0Amazon148
all 241 timed
First token

Wait before the first token, median.

  1. 01Granite 4.2 8BIBM Granite148ms
  2. 02LFM3-1.7BLiquidAI152ms
  3. 03LFM2.5-24B-A2BLiquidAI158ms
  4. 04Phi 4Microsoft190ms
  5. 05Llama 3 8B LunarisSao10K204ms
  6. 06Codestral 2508Mistral211ms
  7. 07Mistral Small 3Mistral227ms
  8. 08gpt-oss-safeguard-20bOpenAI244ms
  9. 09Seed-2.0-MiniByteDance Seed263ms
  10. 10Nemotron 3.5 LightningNVIDIA275ms
  11. 11Beam 1.0 501B A23BReflection286ms
  12. 12Llama Guard 4 12BMeta Llama294ms
  13. 13Llama 3.1 8B InstructMeta Llama311ms
  14. 14Nemotron 3 Nano 30B A3BNVIDIA316ms
  15. 15Hermes 4 405BNous329ms
  16. 16Muse Spark OpenMeta337ms
  17. 17gpt-oss-120bOpenAI339ms
  18. 18Llama 4 ScoutMeta Llama355ms
  19. 19Qwen2.5 72B InstructQwen370ms
  20. 20Nova Micro 1.0Amazon371ms
all 241 timed

The top 20 of 241, 438 down to 148 in output tokens per second. Every column stands on zero, so a column half as tall is half the rate. OpenRouter timed its own routed traffic over a 30-minute window and published the percentiles per model; 324 of 359 models carried a figure, and 241 of those cleared 250 requests and are ranked here. A model below that floor keeps its figure on its own page, where the request count sits beside it, but a good half hour over eleven requests is not a ranking. Unlike every other board on this page, these are not scores of a model: the same weights behind a different provider answer at a different speed, so a model's page carries every route separately. Read .

06 / 13

Ling 3.1 Flash: 439 index points per blended dollar.

  1. 01Ling 3.1 FlashinclusionAI43929.6 at $0.068 blended per M
  2. 02MiniMax M3.1 FlashMiniMax22229.1 at $0.131 blended per M
  3. 03DeepSeek V4.1 Flash 1017DeepSeek18828.6 at $0.152 blended per M
  4. 04Claude Haiku 5.5Anthropic18336.5 at $0.20 blended per M
  5. 05MiMo-V2.6-FlashXiaomi15326.7 at $0.175 blended per M
  6. 06GPT-6 LunaOpenAI13927.9 at $0.20 blended per M
  7. 07GLM 5.3 FlashZ.ai11727.8 at $0.238 blended per M
  8. 08Solar Pro 4Upstage10728.2 at $0.262 blended per M
  9. 09GLM 5.3Z.ai7931.7 at $0.402 blended per M
  10. 10Inkling 2 SmallThinking Machines6430.6 at $0.475 blended per M
  11. 11Granite 4.2 8BIBM Granite625.1 at $0.082 blended per M
  12. 12Nemotron 3.5 Super VLNVIDIA5713.6 at $0.238 blended per M
  13. 13Hy4Tencent5534.6 at $0.625 blended per M
  14. 14MiMo-V2.6-ProXiaomi5529.9 at $0.544 blended per M
  15. 15Nemotron 3.5 LightningNVIDIA545.5 at $0.103 blended per M
  16. 16GPT-5.6 LunaOpenAI5323.8 at $0.45 blended per M
  17. 17Beam 1.0 501B A23BReflection3717.7 at $0.475 blended per M
  18. 18MiniMax M3MiniMax3618.9 at $0.525 blended per M
  19. 19Qwen3.8 27BQwen3520.1 at $0.581 blended per M
  20. 20GPT-6.1 TerraOpenAI2651.3 at $2.00 blended per M

Intelligence Index divided by the blended OpenRouter list price per million tokens, 3 input tokens to 1 output. 57 priced models have both figures. 1 scored listing is free and left out: a zero price makes the ratio undefined, not infinite.

07 / 13

Muse Wonder leads websites in the models arena at 1378 Elo.

PositionModelEloWin rateIntelligence
01Muse Wonder137863.8%43.6
02Muse Spark 1.3135959.3%36.6
03Claude Fable 5.2135060.4%64.6
04Kimi K3133759.5%17.4
05Gemini 4 Hydrogen133059.1%53.9
06Claude Opus 5.5132555.6%59.1
07Kimi K3.1 Pro131957.5%46.5
08Gemini 4 Argon131655.5%50.4
09GPT-6.1 Sol131554.0%54.6
10Gemini 4 Carbon131155.5%60.0
11MiMo-V2.6-Pro131051.9%29.9
12Muse Spark 1.2130852.6%not scored
13Claude Sonnet 5.5130653.1%54.5
14Muse Spark Open130551.7%24.9
15Claude Fable 5.1130455.2%50.0
16Claude Opus 5130356.1%44.2
17Gemini 3.7 Flash129453.6%not scored
18DeepSeek V4.1 Flash 1017129351.6%28.6
19Gemini 3.6 Flash129154.2%not scored
20Gemini 3.8 Flash129149.5%30.8
21GLM 5.1129055.1%not scored
22Claude Fable 5128756.3%not scored
23Claude Opus 4.6128261.1%not scored
24GLM 5.3127852.5%31.7
25Claude Sonnet 4.6127660.0%not scored

126 models ranked for websites in the models arena; Elo from pairwise human votes at Design Arena, win rate the share of matchups won. Position is the place in this catalog; the author column and Design Arena's own rank, which counts models the catalog does not carry, are on the full board. 101 more on the full board.

08 / 13

4k to 2M: 34 window sizes across 359 models.

Each tick is one context window the catalog lists, on a logarithmic scale, with how many models carry it. Every window below opens to the models that carry it, widest first.

  1. 2M6 models

    Auto CodeAuto RouterAuto Router (Beta)Grok 4.20Grok 4.20 Multi-AgentPareto Code Router

  2. 1.31M1 model

    Llama 4 Scout

  3. 1.05M85 models

    Beam 1.0 501B A23BDeepSeek V4 Flash 0423DeepSeek V4 Flash 0731DeepSeek V4 Flash Vision ExpDeepSeek V4 Pro 0423DeepSeek V4 Pro 0813DeepSeek V4.1 FlashDeepSeek V4.1 Flash 1017Ember-1GLM 5.2GLM 5.3GLM 5.3 FlashGLM 5.3 FlashXGLM 5.3 FlashX LegalGPT-4.1GPT-4.1 MiniGPT-4.1 NanoGPT-5.4GPT-5.4 ProGPT-5.5GPT-5.5 ProGPT-5.6 LunaGPT-5.6 Luna ProGPT-5.6 SolGPT-5.6 Sol ProGPT-5.6 TerraGPT-5.6 Terra ProGPT-6 AstraGPT-6 Astra ProGPT-6 LunaGPT-6 Luna ProGPT-6 SolGPT-6 Sol ProGPT-6.1 SolGPT-6.1 Sol ProGPT-6.1 TerraGPT-6.1 Terra ProGemini 2.5 FlashGemini 2.5 Flash LiteGemini 2.5 ProGemini 2.5 Pro Preview 06-05Gemini 3 Flash PreviewGemini 3.1 Flash LiteGemini 3.1 Flash Lite PreviewGemini 3.1 Pro PreviewGemini 3.1 Pro Preview Custom ToolsGemini 3.5 FlashGemini 3.5 Flash LiteGemini 3.6 FlashGemini 3.7 FlashGemini 3.8 FlashGemini 4 ArgonGemini 4 CarbonGemini 4 HydrogenHy4Hy4 flashHy4 previewInkling 2 SmallKimi K3Kimi K3.1 ProLing 3.2 Flash (free)LongCat 2.0Lyria 3 Clip PreviewLyria 3 Pro PreviewMiMo-V2.6-FlashMiMo-V2.6-ProMiMo-V2.6-Pro-UltraSpeedMiniMax M3MiniMax M3.1 FlashMistral Large 4Muse EmberMuse Spark 1.1Muse Spark 1.2Muse Spark 1.2 ContributorMuse Spark 1.3Muse Spark 1.3 ContributorMuse Spark OpenMuse WonderNemotron 3.5 LightningNemotron 3.5 Super VLPareto 26.10Pareto 26.10 PreviewParticle-1Qwen3.8 2.4T A95BRipple 65B Preview

  4. 1.04M1 model

    Palmyra X5

  5. 1.02M5 models

    ERNIE-5.5-Flash-Preview (free)ERNIE-Light-DT-1.0MiMo-V2.5-ProQwen3.8 27BUnslopNemo 12B

  6. 1M43 models

    Claude Fable 5Claude Fable 5.1Claude Fable 5.2Claude Haiku 5.5Claude Opus 4.6Claude Opus 4.7Claude Opus 4.8Claude Opus 5Claude Opus 5.5Claude Sonnet 4.5Claude Sonnet 4.6Claude Sonnet 5Claude Sonnet 5.5ExtractFugu MaxFugu UltraFugu Ultra v2FusionGLM 5.3 PrimeGrok 4.3Jev RouterLing 3.1 FlashMiMo-V2.5Nova 2 LiteNova Premier 1.0Qwen-PlusQwen3 Coder FlashQwen3.5 Plus 2026-02-15Qwen3.5 Plus 2026-04-20Qwen3.5-FlashQwen3.6 FlashQwen3.6 PlusQwen3.7 FlashQwen3.7 MaxQwen3.7 PlusQwen3.8 FlashQwen3.8 Max (0902)Qwen3.8 Max PrimeQwen3.8 Omni FlashQwen4 MaxStep 5 PreviewSummarizeSwitchyard

  7. 524k5 models

    InklingInkling SmallLlama 4 MaverickSolar Mini 4Solar Pro 4

  8. 512k1 model

    Dots3-Note Preview (free)

  9. 500k3 models

    Grok 4.5Grok 4.6Grok 4.7

  10. 400k17 models

    GPT Chat LatestGPT-5GPT-5 ImageGPT-5 Image MiniGPT-5 MiniGPT-5 NanoGPT-5 ProGPT-5.1GPT-5.1-CodexGPT-5.1-Codex-MaxGPT-5.1-Codex-MiniGPT-5.2GPT-5.2 ProGPT-5.2-CodexGPT-5.3-CodexGPT-5.4 MiniGPT-5.4 Nano

  11. 300k2 models

    Nova Lite 1.0Nova Pro 1.0

  12. 272k1 model

    GPT-5.4 Image 2

  13. 262k56 models

    Aion 3.5Aion 3.5 MiniDevstral 2 2512Flint (free)Gemma 4 26B A4BGemma 4 31BGlimpseHy3Hy3 previewKimi K2 0905Kimi K2 ThinkingKimi K2.5Kimi K2.6Kimi K2.7 CodeLFM3-1.7BLing 3.0 FlashLing 3.0 TinyMi:dm K 3.0 ProMinistral 3 14B 2512Ministral 3 8B 2512Mistral Large 3 2512Mistral Medium 3.5Mistral Small 4Morph V3 FastMorph V3 LargeMotif 3Nemotron 3 Nano 30B A3BNemotron 3 SuperNemotron 3 UltraNex-N2.5-MiniNex-N2.5-ProNex-N2.6Pareto 26.9Qwen3 235B A22B Instruct 2507Qwen3 30B A3B Instruct 2507Qwen3 Coder 30B A3B InstructQwen3 Coder 480B A35BQwen3 Coder NextQwen3 Next 80B A3B InstructQwen3 VL 235B A22B InstructQwen3 VL 30B A3B InstructQwen3 VL 8B InstructQwen3.5 397B A17BQwen3.5-122B-A10BQwen3.5-27BQwen3.5-35B-A3BQwen3.5-9BQwen3.6 27BQwen3.6 35B A3BSakana NamazuSeed 2.1 TurboSeed-2.0-LiteSeed-2.0-MiniStep 3.7 FlashTernary Bonsai 2 27BTernary Bonsai 2 9B

  14. 256k8 models

    Codestral 2508Command AGrok Build 0.1Mistral Small 3.2 24BNemotron 3 Nano Omni (free)North Mini Code (free)Relace Apply 3Relace Search

  15. 205k4 models

    GLM 4.6GLM 4.7GLM 5GLM 5.1

  16. 203k2 models

    GLM 5 TurboGLM 5V Turbo

  17. 200k11 models

    Claude Haiku 4.5Claude Opus 4.1Claude Opus 4.5Free Models RouterGLM 4.7 FlashSonar ProSonar Pro Searcho3o3 Mini (batch)o3 Proo4 Mini (batch)

  18. 198k1 model

    MiniMax M2.5

  19. 197k1 model

    MiniMax M2.7

  20. 192k1 model

    Command A+

  21. 164k8 models

    DeepSeek V3DeepSeek V3 0324DeepSeek V3.1DeepSeek V3.1 TerminusDeepSeek V3.2DeepSeek V3.2 ExpLlama Guard 4 12BR1 0528

  22. 131k34 models

    Aion-2.0Aion-3.0Aion-3.0-MiniCydonia 24B V4.1GLM 4.5GLM 4.5 AirGLM 4.6VGemini Omni 1.1 FlashGemini Omni 1.2 FlashGemma 3 12BGemma 3 27BGemma 3 4BGranite 4.2 8BHermes 4 405BKimi K2 0711Ling 3.0 Flash VLLlama 3.1 70B InstructLlama 3.1 8B InstructLlama 3.1 Euryale 70B v2.2Llama 3.2 3B InstructLlama 3.3 70B InstructMinistral 3 3B 2512Mistral Large 2407Mistral Medium 3Mistral Medium 3.1Mistral NemoMuse Glimmer 30BNano Banana 2 (Gemini 3.1 Flash Image)Nano Banana Pro (Gemini 3 Pro Image)Nemotron 3.5 Content SafetyQwen3 32Bgpt-oss-120bgpt-oss-20bgpt-oss-safeguard-20b

  23. 128k26 models

    Body Builder (beta)Command R (08-2024)Command R+ (08-2024)Command R7B (12-2024)GPT AudioGPT Audio MiniGPT-4 Turbo (batch)GPT-4oGPT-4o (2024-05-13)GPT-4o (2024-08-06)GPT-4o (2024-11-20)GPT-4o-miniGPT-4o-mini (2024-07-18)GPT-5.2 ChatLFM2.5-24B-A2BMistral LargeMistral Small 3.1 24BNova Micro 1.0Qwen2.5 VL 72B InstructQwen3 235B A22B Thinking 2507Schematron V2 SmallSchematron V2 TurboSonar Deep ResearchSonar Reasoning ProUI-TARS 7BUncensored

  24. 127k1 model

    Sonar

  25. 66k7 models

    GLM 4.5VMixtral 8x22B InstructNano Banana 2 (Gemini 3.1 Flash Image Preview)Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Nano Banana 2.1Nano Banana Pro (Gemini 3 Pro Image Preview)Reka Flash 3

  26. 60k1 model

    Llama 3.2 1B Instruct

  27. 41k2 models

    Qwen3 14BQwen3 30B A3B

  28. 37k1 model

    Perceptron Mk1.5

  29. 33k10 models

    Aion-RP 1.0 (8B)Magnum v4 72BMistral Small 3Nano Banana (Gemini 2.5 Flash Image)Perceptron Mk1Qwen2.5 72B InstructQwen2.5 Coder 32B InstructSabaSkyfall 36B V2Voxtral Small 24B 2507

  30. 32k2 models

    Moondream 3 PreviewMoondream 3.1

  31. 16k3 models

    GPT-3.5 Turbo 16kPhi 4Reka Edge

  32. 8k8 models

    GPT-4Gemma 2 27BHy-MT2-1.8BHy-MT2-30B-A3BHy-MT2-7BLlama 3 8B LunarisMythoMax 13BWeaver (alpha)

  33. 6k1 model

    ReMM SLERP 13B

  34. 4k1 model

    GPT-3.5 Turbo (older v0613)

09 / 13

21 models listed in Nov; 27 in the last thirty days.

  1. Qwen4 MaxQwennot scored
  2. Hy4 flashTencentnot scored
  3. Ling 3.2 Flash (free)inclusionAInot scored
  4. Gemini 4 HydrogenGoogleintelligence 53.9
  5. ERNIE-5.5-Flash-Preview (free)Baidunot scored
  6. ERNIE-Light-DT-1.0Baidunot scored
  7. Gemini 4 CarbonGoogleintelligence 60.0
  8. Nex-N2.6Nex AGInot scored
  9. Claude Fable 5.2Anthropicintelligence 64.6
  10. Muse EmberMetanot scored
  11. Inkling 2 SmallThinking Machinesintelligence 30.6
  12. LFM3-1.7BLiquidAInot scored
  13. Gemini Omni 1.2 FlashGooglenot scored
  14. Mi:dm K 3.0 ProKTnot scored
  15. Auto RouterOpenRouternot scored
  16. Auto CodeOpenRouternot scored
  17. Gemini Omni 1.1 FlashGooglenot scored
  18. Ripple 65B PreviewRicursivenot scored
  19. Hy4Tencentintelligence 34.6
  20. Motif 3Motif Technologiesnot scored
  21. GLM 5.3 FlashX LegalZ.ainot scored
  1. Particle-1Particle AInot scored
  2. GPT-6.1 TerraOpenAIintelligence 51.3
  3. GPT-6.1 Terra ProOpenAInot scored
  4. Kimi K3.1 ProMoonshotAIintelligence 46.5
  5. Flint (free)Recursivenot scored
  6. MiniMax M3.1 FlashMiniMaxintelligence 29.1
  7. DeepSeek V4.1 Flash 1017DeepSeekintelligence 28.6
  8. GlimpseEloriannot scored
  9. Ternary Bonsai 2 9BPrismMLnot scored

The 30 newest listings, by the date OpenRouter listed them, which is not always the date the provider released them.

10 / 13

OpenAI lists 61 of 359; 61 authors in all.

Share of the ranked catalog by author, the 8 largest and the rest together. This is what OpenRouter lists, not what anyone uses.

11 / 13

98 of the 186 models sold by more than one company cost at least twice as much from the dearest seller as the cheapest; the median is 2x.

  1. 01GLM 5.2Z.ai330x30 sellers · $0.007 to $2.31
  2. 02DeepSeek V4.1 FlashDeepSeek30.12x37 sellers · $0.0498 to $1.50
  3. 03GLM 5.3Z.ai30.04x47 sellers · $0.0699 to $2.10
  4. 04Kimi K3MoonshotAI29.41x29 sellers · $0.153 to $4.50
  5. 05Qwen3.8 27BQwen28.29x24 sellers · $0.035 to $0.99
  6. 06DeepSeek V4.1 Flash 1017DeepSeek20x51 sellers · $0.03 to $0.60
  7. 07DeepSeek V4 Pro 0813DeepSeek19.82x21 sellers · $0.0999 to $1.98
  8. 08Kimi K3.1 ProMoonshotAI18.31x28 sellers · $0.24 to $4.40
  9. 09DeepSeek V4 Flash 0423DeepSeek15.22x13 sellers · $0.0138 to $0.21
  10. 10GPT-6 AstraOpenAI12x6 sellers · $5.00 to $60.00
  11. 11GPT-6 Astra ProOpenAI12x5 sellers · $5.00 to $60.00
  12. 12GPT-6.1 SolOpenAI12x6 sellers · $1.00 to $12.00
  13. 13GPT-6.1 Sol ProOpenAI12x5 sellers · $1.00 to $12.00
  14. 14gpt-oss-20bOpenAI11.76x12 sellers · $0.017 to $0.20
  15. 15gpt-oss-120bOpenAI11.67x23 sellers · $0.03 to $0.35
  16. 16Llama 3.1 8B InstructMeta Llama11x4 sellers · $0.02 to $0.22
  17. 17DeepSeek V4 Flash 0731DeepSeek10x26 sellers · $0.044 to $0.44
  18. 18Mistral NemoMistral9.06x6 sellers · $0.0166 to $0.15
  19. 19Gemma 3 12BGoogle8.75x4 sellers · $0.04 to $0.35
  20. 20Gemma 4 31BGoogle7.5x18 sellers · $0.10 to $0.75

The same weights, sold by several companies at different prices: each bar is how much dearer the dearest seller is than the cheapest, per million input tokens. The 20 widest of the 186 are drawn, and 25 of them are 5x or wider. The bars are linear, so a bar twice as long is twice the ratio. Unlike every other board on this page, this ranks nothing about a model: two models with the same spread are not equally good, and a wide spread means only that which company serves the request is worth deciding. Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. Read .

12 / 13

DeepSeek V4.1 Flash 1017 ran 23.9% of the tokens OpenRouter counted over the 30 days to 20 Nov 2026, and the top 50 models of each day ran 94.4% of them.

  1. 01DeepSeek V4.1 Flash 1017DeepSeek · in the top 50 on 30 of 30 days23.9%231T
  2. 02Hy4 flashTencent · in the top 50 on 5 of 30 days9.6%92.7T
  3. 03DeepSeek V4.1 FlashDeepSeek · in the top 50 on 30 of 30 days7.2%69.3T
  4. 04GLM 5.3 FlashZ.ai · in the top 50 on 30 of 30 days5.9%56.8T
  5. 05GPT-6 LunaOpenAI · in the top 50 on 30 of 30 days3.8%37.0T
  6. 06Hy4Tencent · in the top 50 on 17 of 30 days3.8%36.7T
  7. 07Claude Opus 5.5Anthropic · in the top 50 on 30 of 30 days3.2%31.0T
  8. 08Step 5 PreviewStepFun · in the top 50 on 29 of 30 days2.7%25.7T
  9. 09MiniMax M3.1 FlashMiniMax · in the top 50 on 29 of 30 days1.9%18.4T
  10. 10Hy4 previewTencent · in the top 50 on 17 of 30 days1.8%17.3T
  11. 11GLM 5.3Z.ai · in the top 50 on 30 of 30 days1.8%17.2T
  12. 12GPT-6.1 SolOpenAI · in the top 50 on 30 of 30 days1.6%15.3T
  13. 13MiMo-V2.6-FlashXiaomi · in the top 50 on 30 of 30 days1.6%15.1T
  14. 14Nemotron 3 Ultra (free)NVIDIA · in the top 50 on 30 of 30 days1.5%14.9T
  15. 15DeepSeek V4 Flash 0731DeepSeek · in the top 50 on 30 of 30 days1.5%14.6T
  16. 16Kimi K3.1 ProMoonshotAI · in the top 50 on 29 of 30 days1.4%13.5T
  17. 17typesafe/jev-1.13-20260917typesafe · in the top 50 on 30 of 30 days1.3%12.8T
  18. 18Claude Haiku 5.5Anthropic · in the top 50 on 30 of 30 days1.3%12.1T
  19. 19Muse Spark 1.3 ContributorMeta · in the top 50 on 30 of 30 days1.2%12.0T
  20. 20GPT-5.6 LunaOpenAI · in the top 50 on 30 of 30 days1.2%11.2T

The apps sending the tokens, 12 of the 50 OpenRouter ranks

  1. 01Hermes Agent55.9T498M requests
  2. 02Claude Code45.9T440M requests
  3. 03Kilo Code25.6T243M requests
  4. 04pi20.4T194M requests
  5. 05Codex15.3T234M requests
  6. 06OpenClaw13.8T106M requests
  7. 07omp10.4T92.8M requests
  8. 08Strix9.69T89.1M requests
  9. 09DeepSeek Harness9.57T59.3M requests
  10. 10Portkey AI6.46T448M requests
  11. 11Freebuff4.05T36.0M requests
  12. 12Minds by Authentic Intelligence3.88T105M requests

Share of every token OpenRouter counted from 22 Oct 2026 to 20 Nov 2026, 967T in all, including the traffic of every model outside the top 50 of each day, which OpenRouter publishes as one row. A model ranks by its tokens over the whole window; the line beside it is its share of each day on its own scale from zero, and a gap is a day it was outside the top 50 rather than a day it ran nothing. Token totals are prompt plus completion tokens as each provider tokenizer counts them, so a count on one model does not compare directly with a count on another. They are traffic through OpenRouter only, not the whole market: a model sold mostly through its maker directly can be large and rank low here. A ranked model the catalog does not list is named by the id OpenRouter ranks it under, without a link. Source: OpenRouter (openrouter.ai/rankings), as of 2026-11-21T16:13:12.792Z. CC BY 4.0. Source: OpenRouter (openrouter.ai/apps), as of 2026-11-21T16:13:13.837Z. CC BY 4.0. Read .

13 / 13

How these rankings are measured.

Four plates, read left to right: what was measured, when it was read, what it does not establish, and where every figure came from.

  1. 01What is measured.

    Three Artificial Analysis indexes, each 0 to 100, and Design Arena's Elo per board, exactly as OpenRouter's catalog carries them; OpenRouter's own timings of the traffic it routes; and what each company selling a model asks for it. The usage board is not a measurement: it is the tokens each model and app ran through OpenRouter, as OpenRouter counts them. Vitna re-ran none of them and adjusted nothing. A ranking counts one row per model; OpenRouter's batch and free listings of the same model are not ranked twice.

  2. 02Freshness.

    The catalog was read 17 Nov 16:29 UTC and this page is rebuilt when a read changes a figure. The catalog does not carry the dates the indexes were run, so a score's own date is not printed here.

  3. 03What they do not establish.

    Not quality from adoption: the usage board says how much a model ran through one router, and no other board uses it. Not Vitna's evaluation: the engine scores quality on your own representative cases, and none of that is here. A third-party index is a third party's opinion, dated.

  4. 04Where the figures come from.

    Intelligence Index, Coding Index, Agentic Index from Artificial Analysis; Elo, win rate and rank from Design Arena; both as republished in OpenRouter's public model list, cited below with the sha256 of the document read. Throughput and first-token times from OpenRouter's own model pages, per-seller prices from its public endpoints API, and usage from its keyed rankings datasets under CC BY 4.0, each mirrored on its own schedule and dated where it is drawn.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Vitna ran none of these evaluations and adjusted none of these figures.