Model intelligence
678 AI models
Capability (Epoch Capabilities Index and individual benchmarks from Epoch AI), API list prices and context windows (OpenRouter), release dates and open-weight status. Benchmarks are shown individually - never merged into one invented score.
Epoch Capabilities Index by release date
Capability frontier
Source: Epoch AI (CC BY 4.0). ECI is Epoch's item-response model over many benchmarks; each model page shows its 90% confidence interval.
Blended $/1M tokens (3:1 input:output) vs ECI
Price vs capability
Only models with both an ECI score and an OpenRouter price. Normalisation: 75% input + 25% output price. Click a point to open the model.
| # | Model | ECI ↓ | Context | In $/M | Weights | Released |
|---|---|---|---|---|---|---|
| 201 | Mixtral 8x7B | 118.5 | — | — | Open source | 11 Dec 2023 |
| 202 | Claude 3 Haiku | 118.4 | — | — | Closed | 7 Mar 2024 |
| 203 | Ministral 3B | 118.1 | — | — | Closed | 16 Oct 2024 |
| 204 | Yi-34B 01.AI | 117.4 | — | — | Open source | 2 Nov 2023 |
| 205 | phi-3-mini 3.8B | 117.4 | — | — | Open source | 23 Apr 2024 |
| 206 | Stable Beluga 2 | 117.1 | — | — | Open source | 20 Jul 2023 |
| 207 | Gemini 1.0 Pro | 117.0 | — | — | Closed | 13 Dec 2023 |
| 208 | Llama 3.1-8B | 116.6 | — | — | Open source | 23 Jul 2024 |
| 209 | Llama 3-8B | 116.5 | — | — | Open source | 18 Apr 2024 |
| 210 | Qwen2.5-Coder-14B | 116.4 | — | — | Closed | 18 Sep 2024 |
| 211 | Gemma 3 4B | 116.0 | 131K | $0.050 | Open source | 12 Mar 2025 |
| 212 | GPT-3.5 Turbo (Jan 2024) | 115.7 | — | — | Closed | 25 Jan 2024 |
| 213 | PaLM 2-L | 115.0 | — | — | Closed | 17 May 2023 |
| 214 | Llama 2-70B | 113.8 | — | — | Open source | 18 Jul 2023 |
| 215 | GPT-3.5 Turbo (Jun 2023) | 113.4 | — | — | Closed | 13 Jun 2023 |
| 216 | Qwen2.5-Coder (7B) | 113.2 | — | — | Open source | 18 Sep 2024 |
| 217 | Qwen-14B | 113.0 | — | — | Open source | 24 Sep 2023 |
| 218 | Mistral 7B v0.1 | 112.2 | — | — | Open source | 27 Sep 2023 |
| 219 | Falcon-180B | 112.2 | — | — | Open source | 6 Sep 2023 |
| 220 | internlm-20b | 112.1 | — | — | Closed | 18 Sep 2023 |
| 221 | Gemma 7B | 112.0 | — | — | Open source | 21 Feb 2024 |
| 222 | DeepSeek LLM 67B | 110.6 | — | — | Open source | 29 Nov 2023 |
| 223 | LLaMA-65B | 110.2 | — | — | Open source | 24 Feb 2023 |
| 224 | Falcon 2 11B | 109.6 | — | — | Open source | 9 May 2024 |
| 225 | Mistral 7B v0.3 | 109.0 | — | — | Open source | 27 May 2024 |
| 226 | DeepSeek-Coder-V2-Lite-Base | 108.9 | — | — | Closed | 13 Jun 2024 |
| 227 | PaLM 2-M | 108.3 | — | — | Closed | 17 May 2023 |
| 228 | Phi-2 | 108.0 | — | — | Open source | 12 Dec 2023 |
| 229 | Qwen2.5-Coder-3B | 107.8 | — | — | Closed | 18 Sep 2024 |
| 230 | Nemotron-4 15B | 107.7 | — | — | Closed | 26 Feb 2024 |
| 231 | Yi-9B | 107.6 | — | — | Closed | 1 Mar 2024 |
| 232 | LLaMA-33B | 107.4 | — | — | Open source | 24 Feb 2023 |
| 233 | Qwen-7B | 106.8 | — | — | Open source | 28 Sep 2023 |
| 234 | PaLM 2-S | 106.2 | — | — | Closed | 17 May 2023 |
| 235 | Llama 2-13B | 106.2 | — | — | Open source | 18 Jul 2023 |
| 236 | Llama 2-34B | 105.2 | — | — | Closed | 18 Jul 2023 |
| 237 | StarCoder 2 15B | 105.0 | — | — | Open source | 20 Feb 2024 |
| 238 | Yi 6B 01.AI | 104.8 | — | — | Open source | 22 Nov 2023 |
| 239 | Falcon-40B | 104.5 | — | — | Open source | 25 May 2023 |
| 240 | Baichuan2-13B Baichuan | 103.2 | — | — | Open source | 6 Sep 2023 |
| 241 | Qwen2.5-Coder (1.5B) | 103.0 | — | — | Open source | 18 Sep 2024 |
| 242 | internlm-7b | 102.9 | — | — | Closed | 5 Jul 2023 |
| 243 | Llama 3.2 1B | 102.0 | — | — | Open source | 24 Sep 2024 |
| 244 | INTELLECT-1 Prime Intellect,Hugging Face,Arcee AI | 100.9 | — | — | Closed | 29 Nov 2024 |
| 245 | MPT-30B MosaicML | 100.7 | — | — | Open source | 22 Jun 2023 |
| 246 | LLaMA-13B | 100.6 | — | — | Open source | 24 Feb 2023 |
| 247 | Llama 2-7B | 99.1 | — | — | Open source | 18 Jul 2023 |
| 248 | chatglm2-6b | 99.0 | — | — | Closed | 24 Jun 2023 |
| 249 | LLaMA-7B | 96.7 | — | — | Open source | 24 Feb 2023 |
| 250 | Baichuan 2-7B Baichuan | 96.4 | — | — | Open source | 20 Sep 2023 |
| 251 | DeepSeek Coder 33B | 96.3 | — | — | Open source | 2 Nov 2023 |
| 252 | Falcon-7B | 95.1 | — | — | Open source | 24 Apr 2023 |
| 253 | CodeQwen1.5-7B | 94.9 | — | — | Closed | 15 Apr 2024 |
| 254 | vicuna-13b-v1.1 | 94.7 | — | — | Closed | 12 Apr 2023 |
| 255 | MPT-7B MosaicML | 94.6 | — | — | Open source | 5 May 2023 |
| 256 | Gemma 2B | 94.2 | — | — | Open source | 21 Feb 2024 |
| 257 | StarCoder 2 7B | 93.6 | — | — | Open source | 20 Feb 2024 |
| 258 | XGen-7B Salesforce | 93.4 | — | — | Open source | 27 Jun 2023 |
| 259 | Qwen-1_8B | 93.0 | — | — | Closed | 30 Nov 2023 |
| 260 | open_llama_7b | 91.7 | — | — | Closed | 7 Jun 2023 |
| 261 | Phi-1.5 | 91.6 | — | — | Open source | 11 Sep 2023 |
| 262 | Baichuan1-7B Baichuan | 90.5 | — | — | Open source | 1 Jun 2023 |
| 263 | RedPajama-INCITE-7B-Base | 90.5 | — | — | Closed | 4 May 2023 |
| 264 | Dolly 2.0-12b Databricks | 89.7 | — | — | Open source | 11 Apr 2023 |
| 265 | DeepSeek Coder 6.7B | 89.7 | — | — | Open source | 2 Nov 2023 |
| 266 | StarCoder 2 3B | 88.8 | — | — | Open source | 22 Feb 2024 |
| 267 | Qwen2.5-Coder-0.5B | 88.2 | — | — | Closed | 18 Sep 2024 |
| 268 | Cerebras-GPT-13B | 83.3 | — | — | Open source | 20 Mar 2023 |
| 269 | DeepSeek Coder 1.3B | 63.7 | — | — | Open source | 2 Nov 2023 |
| 270 | stablelm-tuned-alpha-7b | 55.8 | — | — | Closed | 19 Apr 2023 |
| 271 | GPT-6.1 Sol OpenAI· reasoning | — | 1.05M | $2.00 | ? | 29 Sep 2026 |
| 272 | GPT-6 Sol OpenAI· reasoning | — | 1.05M | $2.00 | ? | 22 Sep 2026 |
| 273 | Grok 4.7 xAI· reasoning | — | 500K | $2.00 | ? | 21 Sep 2026 |
| 274 | MiMo-V2.6-Pro Xiaomi· reasoning | — | 1.05M | $0.43 | Open source | 21 Sep 2026 |
| 275 | Qwen3.8 2.4T A95B Alibaba (Qwen)· reasoning | — | 1.05M | $2.00 | Open source | 12 Aug 2026 |
| 276 | MiMo-V2.6-Flash Xiaomi· reasoning | — | 1.05M | $0.14 | Open source | 21 Sep 2026 |
| 277 | GPT-6 Luna OpenAI· reasoning | — | 1.05M | $0.10 | ? | 22 Sep 2026 |
| 278 | Qwen3.8 27B Alibaba (Qwen)· reasoning | — | 1M | $0.42 | Open source | 14 Aug 2026 |
| 279 | Gemini 3.1 Pro Preview Google· reasoning | — | 1.05M | $2.00 | ? | 19 Feb 2026 |
| 280 | MiMo-V2.5-Pro Xiaomi· reasoning | — | 1.05M | $0.43 | Open source | 22 Apr 2026 |
| 281 | Hy3 preview Tencent· reasoning | — | 262K | $0.18 | Open source | 22 Apr 2026 |
| 282 | Grok 4.3 xAI· reasoning | — | 1M | $1.25 | ? | 30 Apr 2026 |
| 283 | Ling 3.0 Flash VL inclusionAI (Ant Group)· reasoning | — | 262K | $0.021 | Open source | 10 Sep 2026 |
| 284 | Ling 3.0 Flash Fin inclusionAI (Ant Group)· reasoning | — | 262K | $0.060 | ? | 27 Aug 2026 |
| 285 | Ling 3.0 Flash inclusionAI (Ant Group)· reasoning | — | 262K | $0.021 | Open source | 23 Jul 2026 |
| 286 | LongCat 2.0 Meituan· reasoning | — | 1.05M | $0.30 | Open source | 20 Jul 2026 |
| 287 | Qwen3.6 35B A3B Alibaba (Qwen)· reasoning | — | 262K | $0.15 | Open source | 27 Apr 2026 |
| 288 | Gemini 2.5 Pro Google· reasoning | — | 1.05M | $1.25 | ? | 17 Jun 2025 |
| 289 | Gemini 3.1 Flash Lite Preview Google· reasoning | — | 1.05M | $0.25 | ? | 3 Mar 2026 |
| 290 | Qwen3.5-122B-A10B Alibaba (Qwen)· reasoning | — | 262K | $0.26 | Open source | 25 Feb 2026 |
| 291 | DeepSeek V3.1 Terminus DeepSeek· reasoning | — | 164K | $0.30 | Open source | 22 Sep 2025 |
| 292 | Command A | — | 256K | $2.50 | Open source | 13 Mar 2025 |
| 293 | Nemotron 3.5 Lightning NVIDIA· reasoning | — | 262K | $0.060 | Open source | 11 Aug 2026 |
| 294 | Nemotron 3 Super NVIDIA· reasoning | — | 262K | $0.080 | Open source | 11 Mar 2026 |
| 295 | Qwen3 235B A22B Thinking 2507 Alibaba (Qwen)· reasoning | — | 131K | $0.23 | Open source | 25 Jul 2025 |
| 296 | Mercury 2.5 Inception Labs· reasoning | — | 260K | $0.040 | ? | 8 Sep 2026 |
| 297 | Mistral Small 4 Mistral AI· reasoning | — | 262K | $0.15 | Open source | 16 Mar 2026 |
| 298 | Granite 4.2 8B IBM· reasoning | — | 131K | $0.060 | Open source | 31 Aug 2026 |
| 299 | o3 Mini High OpenAI· reasoning | — | 200K | $1.10 | ? | 12 Feb 2025 |
| 300 | Trinity Large Thinking Arcee AI· reasoning | — | 262K | $0.25 | Open source | 1 Apr 2026 |
Showing 100 of 678. Benchmark cells are % (best reported setting). 3.9K benchmark results in total.← PrevNext →Export CSV