Epoch Capabilities Index Leaderboard
composite194 models · Updated October 2026
On the Epoch Capabilities Index benchmark, Claude Opus 5.5 ranks #1 with a score of 167, while Mistral Nemo offers the best score-per-dollar at $0.03/1M output tokens. The full ranking, with cost per million tokens, is below.
💰 Best value
Mistral Nemo — score 119 at $0.03/1M output tokens
| # | Model | Score | Output / 1M | Context |
|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 167 | $20.00 | 1M |
| 2 | GPT-6 AstraOpenAI | 166 | $50.00 | 1.1M |
| 3 | GPT-6.1 SolOpenAI | 166 | $10.00 | 1.1M |
| 4 | Claude Fable 5.1Anthropic | 165 | $50.00 | 1M |
| 5 | Gemini 4 ArgonGoogle | 165 | $10.00 | 1M |
| 6 | Claude Sonnet 5.5Anthropic | 164 | $10.00 | 1M |
| 7 | Claude Opus 5Anthropic | 163 | $25.00 | 1M |
| 8 | GPT-6 SolOpenAI | 163 | $10.00 | 1.1M |
| 9 | Claude Fable 5Anthropic | 162 | $50.00 | 1M |
| 10 | GPT-5.5 ProOpenAI | 162 | $180.00 | 1.1M |
| 11 | GPT-5.6 SolOpenAI | 162 | $20.00 | 1.1M |
| 12 | GPT-5.6 TerraOpenAI | 159 | $12.00 | 1.1M |
| 13 | GPT-5.4 ProOpenAI | 159 | $180.00 | 1.1M |
| 14 | Claude Haiku 5.5Anthropic | 159 | $0.50 | 1M |
| 15 | GPT-5.5OpenAI | 159 | $30.00 | 1.1M |
| 16 | Claude Opus 4.8Anthropic | 158 | $25.00 | 1M |
| 17 | Kimi K3MoonshotAI | 157 | $15.00 | 1M |
| 18 | Gemini 3.7 FlashGoogle | 157 | $3.75 | 1M |
| 19 | Gemini 3.8 FlashGoogle | 157 | $3.75 | 1M |
| 20 | Qwen3.8 MaxAlibaba | 157 | $6.00 | 1M |
| 21 | GPT-5.3-CodexOpenAI | 157 | $14.00 | 400K |
| 22 | Muse Spark 1.3Meta | 157 | $4.25 | 1M |
| 23 | GPT-5.4OpenAI | 157 | $15.00 | 1.1M |
| 24 | Claude Sonnet 5Anthropic | 156 | $10.00 | 1M |
| 25 | GPT-5.6 LunaOpenAI | 156 | $1.20 | 1.1M |
| 26 | Claude Opus 4.7Anthropic | 156 | $25.00 | 1M |
| 27 | GPT-6 LunaOpenAI | 156 | $0.50 | 1.1M |
| 28 | Grok 4.6SpaceXAI | 156 | $6.00 | 500K |
| 29 | GPT-5.2 ProOpenAI | 156 | $168.00 | 400K |
| 30 | GLM 5.3Z.ai | 156 | $4.80 | 1M |
| 31 | DeepSeek V4.1 FlashDeepSeek | 156 | $0.89 | 1M |
| 32 | DeepSeek V4 Pro 0423DeepSeek | 155 | $1.91 | 1M |
| 33 | Claude Opus 4.6Anthropic | 155 | $25.00 | 1M |
| 34 | Muse Spark 1.2Meta | 155 | $4.25 | 1M |
| 35 | Gemini-3.1-ProPoe | 155 | $12.12 | 1M |
| 36 | DeepSeek V4 Flash 0423DeepSeek | 154 | $0.18 | 1M |
| 37 | Gemini 3.5 FlashGoogle | 154 | $9.00 | 1M |
| 38 | Grok 4.7SpaceXAI | 154 | $6.00 | 500K |
| 39 | Gemini 3.6 FlashGoogle | 154 | $3.75 | 1M |
| 40 | Muse Spark 1.1Meta | 154 | $4.25 | 1M |
| 41 | Grok 4.5SpaceXAI | 154 | $6.00 | 500K |
| 42 | Qwen3.7 MaxQwen | 154 | $4.43 | 1M |
| 43 | GPT-5.2OpenAI | 153 | $14.00 | 400K |
| 44 | GLM 5.3 FlashZ.ai | 153 | $0.90 | 1M |
| 45 | Gemini 3 ProOpenCode | 153 | $12.00 | 1M |
| 46 | Claude Sonnet 4.6Anthropic | 152 | $15.00 | 1M |
| 47 | Muse SparkMeta AI | 152 | — | — |
| 48 | Grok 4.20SpaceXAI | 152 | $2.50 | 2M |
| 49 | GLM 5.2Z.ai | 152 | $7.00 | 1M |
| 50 | Gemini 3 FlashOpenCode | 152 | $3.00 | 1M |
| 51 | Qwen3.8 27BQwen | 152 | $2.20 | 1M |
| 52 | Kimi K2.6MoonshotAI | 151 | $3.41 | 262K |
| 53 | GPT-5 ProOpenAI | 151 | $120.00 | 400K |
| 54 | Inkling SmallThinking Machines | 150 | $1.20 | 524K |
| 55 | Claude Opus 4.5Anthropic | 150 | $25.00 | 200K |
| 56 | GPT-5OpenAI | 150 | $10.00 | 400K |
| 57 | GLM 5.1Z.ai | 150 | $3.04 | 205K |
| 58 | Kimi K2.7 CodeMoonshotAI | 150 | $3.35 | 262K |
| 59 | GPT-5.1OpenAI | 150 | $10.00 | 400K |
| 60 | GPT-5.4 MiniOpenAI | 149 | $4.50 | 400K |
| 61 | InklingThinking Machines | 149 | $4.05 | 524K |
| 62 | Kimi K2.5MoonshotAI | 148 | $3.00 | 262K |
| 63 | Qwen3.6 PlusQwen | 147 | $1.95 | 1M |
| 64 | o3 ProOpenAI | 147 | $80.00 | 200K |
| 65 | Qwen3.7 PlusAlibaba | 147 | $1.60 | 1M |
| 66 | MiniMax M3MiniMax | 147 | $1.20 | 1M |
| 67 | Mistral Large 4Mistral | 147 | $2.09 | 1M |
| 68 | o3OpenAI | 147 | $8.00 | 200K |
| 69 | Claude Sonnet 4.5Anthropic | 147 | $15.00 | 1M |
| 70 | MiniMax M2.5MiniMax | 147 | $0.95 | 205K |
| 71 | Qwen3.5 397B-A17BAlibaba | 147 | $3.60 | 262K |
| 72 | Qwen3.6 27BQwen | 146 | $3.20 | 262K |
| 73 | Grok-4Poe | 146 | $15.00 | 256K |
| 74 | Qwen3.5 PlusAlibaba | 146 | $2.40 | 1M |
| 75 | GPT-5.4 NanoOpenAI | 146 | $1.25 | 400K |
| 76 | DeepSeek V3.2DeepSeek | 146 | $0.80 | 164K |
| 77 | Kimi K2 0711MoonshotAI | 146 | $2.30 | 131K |
| 78 | MiniMax M2.7MiniMax | 146 | $1.20 | 205K |
| 79 | GLM 5Z.ai | 146 | $3.20 | 205K |
| 80 | Nemotron 3 UltraNVIDIA | 146 | $2.40 | 262K |
| 81 | o4 MiniOpenAI | 146 | $4.40 | 200K |
| 82 | GPT-5 MiniOpenAI | 145 | $2.00 | 400K |
| 83 | Gemini 2.5 Pro (Jun 2025)Google DeepMind | 145 | — | — |
| 84 | Gemini 3.5 Flash LiteGoogle | 145 | $2.50 | 1M |
| 85 | DeepSeek V3.2 ExpDeepSeek | 145 | $0.41 | 164K |
| 86 | Qwen3.7 FlashAlibaba | 145 | $0.13 | 1M |
| 87 | Gemini 3.1 Flash LiteGoogle | 144 | $1.50 | 1M |
| 88 | Grok 4 FastZenMux | 144 | $0.50 | 2M |
| 89 | Qwen3.6 35B-A3BAlibaba | 144 | $1.49 | 262K |
| 90 | Claude Opus 4.1Anthropic | 144 | $75.00 | 200K |
| 91 | Qwen3 235B-A22BAlibaba | 144 | $2.80 | 131K |
| 92 | GLM 4.7Z.ai | 144 | $2.20 | 205K |
| 93 | Qwen3.5 FlashAlibaba | 144 | $0.40 | 1M |
| 94 | Qwen3.6 FlashQwen | 143 | $1.13 | 1M |
| 95 | Gemini 2.5 FlashGoogle | 143 | $2.50 | 1M |
| 96 | Gemma 4 31B ITQVAC | 143 | $0.00 | 262K |
| 97 | Claude Opus 4Anthropic | 143 | $75.00 | 200K |
| 98 | Qwen3.5 35B-A3BAlibaba | 142 | $2.00 | 262K |
| 99 | GPT-5.5 InstantZenMux | 142 | $30.00 | 400K |
| 100 | Claude Haiku 4.5Anthropic | 142 | $5.00 | 200K |
| 101 | Qwen3 MaxAlibaba | 142 | $6.00 | 262K |
| 102 | o1OpenAI | 142 | $60.00 | 200K |
| 103 | Gemma 4 26B A4BGoogle | 142 | $0.26 | 262K |
| 104 | Claude Sonnet 4Anthropic | 142 | $15.00 | 200K |
| 105 | Mistral Medium 3.5Mistral | 141 | $7.50 | 262K |
| 106 | DeepSeek-R1Azure | 141 | $5.40 | 164K |
| 107 | Claude 3.7 SonnetHelicone | 141 | $15.00 | 200K |
| 108 | o3 MiniOpenAI | 140 | $4.40 | 200K |
| 109 | Grok 3 MiniPoe | 140 | $0.50 | 131K |
| 110 | gpt-oss-120b (batch)OpenAI | 140 | $0.14 | 131K |
| 111 | DeepSeek V3.1DeepSeek | 140 | $0.95 | 164K |
| 112 | Qwen3 30B A3BQwen | 140 | $0.50 | 41K |
| 113 | Qwen3.5-9BQwen | 139 | $0.15 | 262K |
| 114 | GPT-5 NanoOpenAI | 139 | $0.40 | 400K |
| 115 | Qwen3 235B A22B Instruct 2507Qwen | 139 | $0.55 | 262K |
| 116 | Qwen3 32BQwen | 138 | $0.28 | 131K |
| 117 | Grok 3Poe | 138 | $15.00 | 131K |
| 118 | Qwen3 14BAlibaba | 138 | $1.40 | 131K |
| 119 | gpt-oss-20b (batch)OpenAI | 138 | $0.11 | 131K |
| 120 | QwQ 32BAlibaba (China) | 138 | $0.86 | 131K |
| 121 | DeepSeek R1 Distill Qwen 32BAlibaba (China) | 137 | $0.86 | 33K |
| 122 | Qwen3 30B A3B Instruct 2507Qwen | 137 | $0.30 | 262K |
| 123 | GPT-4.1OpenAI | 137 | $8.00 | 1M |
| 124 | GPT-4.5OpenAI | 137 | — | — |
| 125 | Qwen3 8BAlibaba | 136 | $0.70 | 131K |
| 126 | DeepSeek V3DeepSeek | 136 | $0.28 | 164K |
| 127 | o1-miniAzure | 136 | $4.40 | 128K |
| 128 | DeepSeek R1 Distill Qwen 14BAlibaba (China) | 135 | $0.43 | 33K |
| 129 | Gemini 2.0 FlashGoogle | 135 | $0.40 | 1M |
| 130 | Gemini 2.0 ProGoogle DeepMind | 135 | — | — |
| 131 | GPT-4.1 MiniOpenAI | 135 | $1.60 | 1M |
| 132 | o1-previewOpenAI | 135 | — | — |
| 133 | Mistral Medium 3Mistral | 134 | $2.00 | 131K |
| 134 | Gemini 2.5 Flash LiteGoogle | 134 | $0.40 | 1M |
| 135 | Claude 3.5 SonnetDigitalOcean | 134 | $15.00 | 200K |
| 136 | Qwen 2.5 MaxNanoGPT | 133 | $6.39 | 32K |
| 137 | Llama 4 MaverickMeta | 132 | $0.65 | 1M |
| 138 | Mistral Small 3.2Mistral | 132 | $0.30 | 128K |
| 139 | Gemini 1.5 ProGoogle DeepMind | 132 | — | — |
| 140 | Magistral Small 1.2Amazon Bedrock | 131 | $1.50 | 128K |
| 141 | Grok-2xAI | 130 | — | — |
| 142 | Phi 4Microsoft | 130 | $0.14 | 16K |
| 143 | Gemma 3 27BGoogle | 130 | $0.45 | 131K |
| 144 | Llama 4 ScoutMeta | 130 | $0.59 | 1.3M |
| 145 | GPT-4.1 NanoOpenAI | 130 | $0.40 | 1M |
| 146 | Gemini 1.5 FlashNanoGPT | 129 | $0.31 | 2M |
| 147 | GPT-4oOpenAI | 129 | $10.00 | 128K |
| 148 | Qwen2.5 72BNanoGPT | 129 | $0.41 | 131K |
| 149 | Llama 3.1-405BMeta AI | 129 | — | — |
| 150 | Qwen2.5-32BAlibaba | 129 | — | — |
| 151 | Mistral Large 2Mistral AI | 129 | — | — |
| 152 | Mistral Small 3.1Azure | 127 | $0.30 | 128K |
| 153 | Llama 3.3 70BVenice AI | 127 | $2.80 | 128K |
| 154 | GPT-4 TurboOpenAI | 127 | $30.00 | 128K |
| 155 | Claude 3.5 HaikuHelicone | 127 | $4.00 | 200K |
| 156 | Mistral Small 3Mistral | 127 | $0.08 | 33K |
| 157 | Claude 3 OpusLLM Gateway | 127 | $75.00 | 200K |
| 158 | GPT-4o-miniOpenAI | 127 | $0.60 | 128K |
| 159 | Llama 3.1 70BWeights & Biases | 126 | $0.80 | 128K |
| 160 | GPT-4OpenAI | 126 | $60.00 | 8K |
| 161 | Qwen2-72BAlibaba | 125 | — | — |
| 162 | amazon--nova-proSAP AI Core | 124 | $2.13 | 300K |
| 163 | Gemma 3 12BGoogle | 123 | $0.15 | 131K |
| 164 | Llama 3-70BMeta AI | 123 | — | — |
| 165 | Gemma 2 27BGoogle | 122 | $0.65 | 8K |
| 166 | Mixtral 8x22BMistral | 122 | $6.00 | 64K |
| 167 | Mistral LargeMistral AI | 122 | $6.00 | 128K |
| 168 | Claude 3 SonnetAnthropic | 121 | — | — |
| 169 | Claude 2Anthropic | 120 | — | — |
| 170 | Qwen 2.5 Coder 32BAbacus | 120 | $0.79 | 128K |
| 171 | Command R (08-2024)Cohere | 119 | $0.60 | 128K |
| 172 | Claude 2.1Anthropic | 119 | — | — |
| 173 | Mistral NemoMistral | 119 | $0.03 | 131K |
| 174 | GPT-3.5 TurboOpenAI | 119 | $1.50 | 16K |
| 175 | Mixtral 8x7BMistral | 119 | $0.70 | 32K |
| 176 | Claude 3 HaikuAnthropic | 118 | $1.25 | 200K |
| 177 | Ministral 3B (latest)Mistral | 118 | $0.04 | 131K |
| 178 | Yi-34B01.AI | 117 | — | — |
| 179 | phi-3-mini 3.8BMicrosoft | 117 | — | — |
| 180 | Gemini 1.0 ProGoogle DeepMind | 117 | — | — |
| 181 | Llama 3.1 8BGroq | 117 | $0.08 | 131K |
| 182 | Llama 3-8BMeta AI | 116 | — | — |
| 183 | Gemma 3 4BGoogle | 116 | $0.10 | 131K |
| 184 | Llama 2-70BMeta AI | 114 | — | — |
| 185 | Falcon-180BTechnology Innovation Institute | 112 | — | — |
| 186 | LLaMA-65BMeta AI | 110 | — | — |
| 187 | Llama 2-13BMeta AI | 106 | — | — |
| 188 | Llama 2-34BMeta AI | 105 | — | — |
| 189 | Falcon-40BTechnology Innovation Institute | 105 | — | — |
| 190 | Llama-3.2-1BPioneer | 102 | $0.10 | 131K |
| 191 | LLaMA-13BMeta AI | 101 | — | — |
| 192 | Llama 2-7BMeta AI | 99.2 | — | — |
| 193 | LLaMA-7BMeta AI | 96.8 | — | — |
| 194 | Qwen2.5-Coder-0.5BPioneer | 88.3 | $0.10 | 33K |
Frequently asked questions
What is the Epoch Capabilities Index benchmark?
Epoch Capabilities Index is a composite used to evaluate and rank large language models. This page lists 194 models by their Epoch Capabilities Index score.
Which model is best on Epoch Capabilities Index?
Claude Opus 5.5 (Anthropic) currently leads the Epoch Capabilities Index leaderboard with a score of 167.
What is the best value model on Epoch Capabilities Index?
Mistral Nemo offers the best score-per-dollar on Epoch Capabilities Index — a score of 119 at $0.03 per 1M output tokens.
Pricing is indicative — confirm with the provider before production use. Updated October 2026.