LLM Leaderboard: Ranked by Epoch Capabilities Index
107 of 841 tracked models with a published Epoch Capabilities Index (ECI) score (Epoch AI, epoch.ai, CC BY 4.0), refreshed daily. Models without a published ECI are excluded rather than shown with a placeholder — see the full pricing table for those.
| Rank | Model | Developer | Released | ECI | Blended $/1M |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | OpenAI | 2026-09-04 | 169.1 | 20.00 |
| 2 | GPT-6.1 Sol | OpenAI | 2026-09-29 | 168.9 | 4.00 |
| 3 | Claude Opus 5.5 | Anthropic | 2026-09-22 | 168.4 | 8.00 |
| 4 | GPT-6.1 TerraNEW | OpenAI | 2026-10-26 | 167.8 | 2.00 |
| 5 | Claude Fable 5.1 | Anthropic | 2026-09-01 | 166.9 | 20.00 |
| 6 | Gemini 4 ArgonNEW | 2026-10-13 | 166.4 | 4.00 | |
| 7 | Claude Sonnet 5.5 | Anthropic | 2026-09-28 | 166.2 | 4.00 |
| 8 | Claude Opus 5 | Anthropic | 2026-07-24 | 164.4 | 10.00 |
| 9 | Claude Fable 5 | Anthropic | 2026-06-09 | 164.2 | 20.00 |
| 10 | GPT-6 Sol | OpenAI | 2026-09-22 | 164.1 | 4.00 |
| 11 | GPT-5.6 Sol | OpenAI | 2026-07-09 | 163.0 | 8.00 |
| 12 | GPT-5.5 Pro | OpenAI | 2026-04-23 | 162.7 | 67.50 |
| 13 | Muse WonderNEW | Meta | 2026-10-12 | 162.0 | 8.00 |
| 14 | GPT-5.6 Terra | OpenAI | 2026-07-09 | 160.9 | 4.50 |
| 15 | Claude Haiku 5.5NEW | Anthropic | 2026-10-07 | 160.2 | 0.20 |
| 16 | GPT-5.4 Pro | OpenAI | 2026-03-05 | 160.2 | 67.50 |
| 17 | GPT-5.5 | OpenAI | 2026-04-23 | 160.0 | 11.25 |
| 18 | Claude Opus 4.8 | Anthropic | 2026-05-28 | 159.9 | 10.00 |
| 19 | DeepSeek V4.1 Flash 1017NEW | DeepSeek | 2026-10-17 | 159.3 | 0.2625 |
| 20 | Gemini 3.8 Flash | 2026-09-02 | 158.6 | 1.50 | |
| 21 | Kimi K3 | Moonshot AI | 2026-07-16 | 158.5 | 6.00 |
| 22 | Gemini 3.7 Flash | 2026-08-13 | 158.4 | 1.50 | |
| 23 | Muse Spark 1.3 | Meta | 2026-09-02 | 158.0 | 2.00 |
| 24 | Grok 4.6 | xAI | 2026-08-12 | 157.9 | 3.00 |
| 25 | GPT-5.4 | OpenAI | 2026-03-05 | 157.8 | 5.625 |
| 26 | Claude Opus 4.7 | Anthropic | 2026-04-16 | 157.6 | 10.00 |
| 27 | GPT-5.2 Pro | OpenAI | 2025-12-11 | 157.2 | 57.75 |
| 28 | GLM-5.3 | Zhipu AI | 2026-08-14 | 156.9 | 1.9421 |
| 29 | GPT-5.6 Luna | OpenAI | 2026-07-09 | 156.8 | 0.45 |
| 30 | Claude Sonnet 5 | Anthropic | 2026-06-30 | 156.7 | 4.00 |
| 31 | GPT-6 Luna | OpenAI | 2026-09-22 | 156.7 | 0.20 |
| 32 | DeepSeek V4.1 Flash | DeepSeek | 2026-09-10 | 156.5 | — |
| 33 | Claude Opus 4.6 | Anthropic | 2026-02-05 | 156.3 | 10.00 |
| 34 | Grok 4.7 | xAI | 2026-09-21 | 156.1 | 3.00 |
| 35 | DeepSeek V4 Pro 0813 | DeepSeek | 2026-08-13 | 155.9 | 0.99 |
| 36 | Qwen3.8 Max 0902 | Alibaba | 2026-09-02 | 155.8 | 3.00 |
| 37 | Muse Spark OpenNEW | Meta | 2026-10-15 | 155.4 | — |
| 38 | Gemini 3.5 Flash | 2026-05-19 | 155.2 | 3.375 | |
| 39 | Gemini 3.6 Flash | 2026-07-21 | 155.2 | 1.50 | |
| 40 | Grok 4.5 | xAI | 2026-07-08 | 155.1 | 3.00 |
| 41 | Muse Spark 1.2 | Meta | 2026-08-05 | 155.1 | 2.00 |
| 42 | Muse Spark 1.1 | Meta | 2026-04-08 | 154.9 | 2.00 |
| 43 | Mistral Large 4 | Mistral | 2026-10-06 | 154.7 | 1.05 |
| 44 | Qwen3.7 Max | Alibaba | 2026-05-21 | 154.5 | 3.75 |
| 45 | GPT-5.2 | OpenAI | 2025-12-11 | 154.4 | 4.8125 |
| 46 | DeepSeek V4 Flash 0731 | DeepSeek | 2026-07-31 | 154.3 | — |
| 47 | GLM-5.3-Flash | Zhipu AI | 2026-08-26 | 153.6 | 0.1942 |
| 48 | GLM-5.2 | Zhipu AI | 2026-06-13 | 153.0 | 1.9421 |
| 49 | Claude Sonnet 4.6 | Anthropic | 2026-02-17 | 152.9 | 6.00 |
| 50 | Claude Opus 4.5 (latest) | Anthropic | 2025-11-24 | 151.8 | 10.00 |
| 51 | Kimi K2.6 | Moonshot AI | 2026-04-20 | 151.6 | 1.7125 |
| 52 | Kimi K2.7 Code | Moonshot AI | 2026-06-12 | 151.2 | 1.7125 |
| 53 | GPT-5 Pro | OpenAI | 2025-10-06 | 151.2 | 41.25 |
| 54 | GPT-5.1 | OpenAI | 2025-11-13 | 150.9 | 3.4375 |
| 55 | Inkling Small | Thinking Machines | 2026-07-30 | 150.8 | — |
| 56 | GLM-5.1 | Zhipu AI | 2026-04-07 | 150.7 | 1.5686 |
| 57 | DeepSeek V4 Pro | DeepSeek | 2026-04-24 | 150.1 | 1.98 |
| 58 | GPT-5 | OpenAI | 2025-08-07 | 150.0 | 3.4375 |
| 59 | GPT-5.4 mini | OpenAI | 2026-03-17 | 149.3 | 1.6875 |
| 60 | Kimi K2.5 | Moonshot AI | 2026-01-27 | 148.7 | 1.20 |
| 61 | Inkling | Thinking Machines | 2026-07-15 | 148.4 | — |
| 62 | Qwen3.5 397B-A17B | Alibaba | 2026-02-15 | 147.9 | 1.35 |
| 63 | Qwen3.7 Plus | Alibaba | 2026-06-02 | 147.8 | 0.70 |
| 64 | MiniMax-M3 | MiniMax | 2026-06-01 | 147.6 | 0.525 |
| 65 | Claude Sonnet 4.5 (latest) | Anthropic | 2025-09-29 | 147.6 | 6.00 |
| 66 | o3-pro | OpenAI | 2025-06-10 | 147.4 | 35.00 |
| 67 | o3 | OpenAI | 2025-04-16 | 147.0 | 3.50 |
| 68 | DeepSeek V4 Flash | DeepSeek | 2026-04-24 | 146.9 | 0.525 |
| 69 | Qwen3.6 27B | Alibaba | 2026-04-22 | 146.7 | 1.35 |
| 70 | GLM-5 | Zhipu AI | 2026-02-12 | 146.4 | 1.1205 |
| 71 | GPT-5.4 nano | OpenAI | 2026-03-17 | 146.4 | 0.4625 |
| 72 | DeepSeek V3.2 | DeepSeek | 2025-12-01 | 146.2 | — |
| 73 | GPT-5 Mini | OpenAI | 2025-08-07 | 146.1 | 0.6875 |
| 74 | Gemini 3.5 Flash Lite | 2026-07-21 | 145.6 | 0.85 | |
| 75 | Kimi K2 Thinking | Moonshot AI | 2025-11-06 | 145.5 | — |
| 76 | o4-mini | OpenAI | 2025-04-16 | 145.2 | 1.925 |
| 77 | Gemini 3.1 Flash Lite | 2026-05-07 | 144.9 | 0.5625 | |
| 78 | Claude Opus 4.1 (latest) | Anthropic | 2025-08-05 | 144.7 | — |
| 79 | Gemma 4 31B IT | 2026-04-02 | 144.2 | — | |
| 80 | Qwen3.5 35B-A3B | Alibaba | 2026-02-23 | 143.9 | 0.6875 |
| 81 | GLM-4.7 | Zhipu AI | 2025-12-22 | 143.2 | 0.859 |
| 82 | Claude Haiku 4.5 (latest) | Anthropic | 2025-10-15 | 142.8 | 2.00 |
| 83 | Qwen3 Max | Alibaba | 2025-09-23 | 142.5 | 2.40 |
| 84 | Mistral Medium 3.5 | Mistral | 2026-04-30 | 141.6 | 3.00 |
| 85 | o1 | OpenAI | 2024-12-05 | 141.2 | 26.25 |
| 86 | DeepSeek V3.1 | DeepSeek | 2025-08-21 | 140.8 | — |
| 87 | o3-mini | OpenAI | 2024-12-20 | 140.6 | 1.925 |
| 88 | GPT OSS 120B | OpenAI | 2025-08-05 | 140.2 | — |
| 89 | Qwen3.5 9B | Alibaba | 2026-02-23 | 139.3 | — |
| 90 | GPT-5 Nano | OpenAI | 2025-08-07 | 139.2 | 0.1375 |
| 91 | DeepSeek-R1 | DeepSeek | 2025-01-20 | 138.6 | — |
| 92 | Qwen3 235B-A22B | Alibaba | 2025-04-28 | 138.4 | 1.225 |
| 93 | Qwen3 32B | Alibaba | 2025-04-28 | 136.9 | 1.225 |
| 94 | QwQ 32B | Alibaba | 2025-03-05 | 136.5 | — |
| 95 | GPT-4.1 | OpenAI | 2025-04-14 | 135.9 | 3.50 |
| 96 | GPT OSS 20B | OpenAI | 2025-08-05 | 135.0 | — |
| 97 | GPT-4.1 mini | OpenAI | 2025-04-14 | 134.3 | 0.70 |
| 98 | Qwen3 30B A3B | Alibaba | 2025-04-28 | 134.3 | — |
| 99 | Mistral Medium 3 | Mistral | 2025-05-07 | 132.8 | 0.80 |
| 100 | Qwen3 8B | Alibaba | 2025-04-29 | 132.2 | 0.31 |
| 101 | DeepSeek-V3 | DeepSeek | 2024-12-26 | 131.8 | — |
| 102 | GPT-4.1 nano | OpenAI | 2025-04-14 | 127.0 | 0.175 |
| 103 | GPT-4o mini | OpenAI | 2024-07-18 | 124.0 | 0.2625 |
| 104 | Mistral Large | Mistral | 2024-02-26 | 120.0 | — |
| 105 | Gemma 2 27B | 2024-07-13 | 118.7 | — | |
| 106 | Mistral Nemo | Mistral | 2024-07-18 | 114.1 | — |
| 107 | Qwen2.5-Coder-0.5B | Alibaba | 2024-11-12 | 68.9 | — |
FAQ
What is the Epoch Capabilities Index (ECI)?
A general-capability benchmark score published by Epoch AI (epoch.ai), CC BY 4.0 licensed. It is Epoch's own composite across multiple public reasoning/knowledge evaluations — we display it as its own named column, we don't blend it with any other benchmark into a further composite of our own.
How often is this leaderboard updated?
Daily. 107 of 841 tracked models currently have a published ECI score; models without one are excluded rather than shown with a placeholder.
How is this different from the model pricing table?
This page ranks purely by ECI. The /models/ page is built for comparing and filtering by price — sort by input, output or blended $/1M tokens, filter by developer or a price ceiling.
Why do some rows show "—" instead of a price?
Neither models.dev nor LiteLLM has published pricing for that model. We never guess a price or display an unpublished price as free — it's shown as "—" instead.
Want the full price breakdown instead of a ranking? See the model pricing table. Want to know what a specific prompt actually costs? Use the token counter.