Benchmarks
Leaderboard
All active models ranked by P50 latency — the median response time for a standard 500-token output, measured from EU (Amsterdam). Green < 500 ms, yellow 500–1000 ms, red > 1000 ms.
Filter:
| # | |||||||
|---|---|---|---|---|---|---|---|
| 1 | pplC | OVH AI Endpoints (GRA) | C | 27 | 450 | — | 2026-06-15 |
| 2 | Mistral-7B-Instruct-v0.3 | OVH AI Endpoints (GRA) | — | 116 | 150 | 97 | 2026-06-15 |
| 3 | Llama-3.1-8B-Instruct | OVH AI Endpoints (GRA) | — | 116 | 147 | 97 | 2026-06-15 |
| 4 | Meta-Llama-3_3-70B-Instruct | OVH AI Endpoints (GRA) | — | 129 | 170 | 97 | 2026-06-15 |
| 5 | Mistral Voxtral Small 24BA | OpenRouter | A | 130 | 137 | — | 2026-06-15 |
| 6 | Qwen2.5-VL-72B-Instruct | OVH AI Endpoints (GRA) | — | 139 | 153 | 95 | 2026-06-15 |
| 7 | Mistral-Small-3.2-24B-Instruct-2506 | OVH AI Endpoints (GRA) | — | 151 | 371 | 100 | 2026-06-15 |
| 8 | Mistral-Nemo-Instruct-2407 | OVH AI Endpoints (GRA) | — | 172 | 400 | 85 | 2026-06-15 |
| 9 | Qwen3.5-397B-A17B | OVH AI Endpoints (GRA) | — | 211 | 223 | 0 | 2026-06-15 |
| 10 | NVIDIA Nemotron Super 49B v1.5A | OpenRouter | A | 247 | 254 | — | 2026-06-15 |
| 11 | Llama 4 MaverickA | OpenRouter | A | 254 | 760 | — | 2026-06-15 |
| 12 | Llama 3.3 70B InstructA | OpenRouter | A | 282 | 329 | — | 2026-06-15 |
| 13 | gpt-oss-20bC | OVH AI Endpoints (GRA) | C | 303 | 333 | 92 | 2026-06-15 |
| 14 | gpt-oss-120bC | OVH AI Endpoints (GRA) | C | 313 | 360 | 100 | 2026-06-15 |
| 15 | gpt-5-chat-latestC | OpenAI | C | 405 | 465 | 99 | 2026-06-15 |
| 16 | o3-miniC | OpenAI | C | 425 | 456 | — | 2026-06-15 |
| 17 | Gemini 2.5 Flash-LiteB | Google Gemini | B | 441 | 454 | 99 | 2026-06-15 |
| 18 | gpt-4.1-nanoC | OpenAI | C | 446 | 604 | 100 | 2026-06-15 |
| 19 | Qwen3-Coder-30B-A3B-Instruct | OVH AI Endpoints (GRA) | — | 449 | 491 | 97 | 2026-06-15 |
| 20 | Qwen3.5-9B | OVH AI Endpoints (GRA) | — | 451 | 532 | 0 | 2026-06-15 |
| 21 | Qwen3-32B | OVH AI Endpoints (GRA) | — | 457 | 491 | 50 | 2026-06-15 |
| 22 | gpt-5.4-nanoC | OpenAI | C | 471 | 524 | — | 2026-06-15 |
| 23 | gpt-4oC | OpenAI | C | 478 | 642 | 98 | 2026-06-15 |
| 24 | Nous Hermes 3 70BA | OpenRouter | A | 516 | 2432 | — | 2026-06-15 |
| 25 | gpt-5.4-miniA | OpenAI | A | 521 | 539 | — | 2026-06-15 |
| 26 | DeepSeek v4 ProA | OpenRouter | A | 579 | 1079 | — | 2026-06-15 |
| 27 | gpt-5C | OpenAI | C | 584 | 659 | — | 2026-06-15 |
| 28 | gpt-5.1-chat-latestC | OpenAI | C | 593 | 822 | — | 2026-06-15 |
| 29 | gpt-5.1B | OpenAI | B | 594 | 689 | — | 2026-06-15 |
| 30 | gpt-4.1B | OpenAI | B | 601 | 678 | 99 | 2026-06-15 |
| 31 | gpt-5.4A | OpenAI | A | 617 | 846 | — | 2026-06-15 |
| 32 | Gemini 2.5 FlashA | Google Gemini | A | 629 | 1084 | 30 | 2026-06-15 |
| 33 | gpt-4o-miniC | OpenAI | C | 638 | 851 | 99 | 2026-06-15 |
| 34 | Qwen 2.5 VL 72B InstructA | OpenRouter | A | 641 | 1169 | — | 2026-06-15 |
| 35 | gpt-5.2-chat-latestC | OpenAI | C | 669 | 816 | — | 2026-06-15 |
| 36 | Claude Haiku 4.5A | Anthropic | A | 691 | 702 | 100 | 2026-06-15 |
| 37 | gpt-5-nanoC | OpenAI | C | 695 | 861 | — | 2026-06-15 |
| 38 | o3C | OpenAI | C | 739 | 1112 | — | 2026-06-15 |
| 39 | gpt-5.3-chat-latestC | OpenAI | C | 747 | 750 | — | 2026-06-15 |
| 40 | Cohere Command-AA | OpenRouter | A | 765 | 1488 | — | 2026-06-15 |
| 41 | Llama 4 ScoutA | OpenRouter | A | 842 | 927 | — | 2026-06-15 |
| 42 | o4-miniC | OpenAI | C | 851 | 1059 | — | 2026-06-15 |
| 43 | Claude Sonnet 4C | Anthropic | C | 861 | 1622 | 100 | 2026-06-15 |
| 44 | Google Lyria 3 Pro PreviewA | OpenRouter | A | 869 | 880 | — | 2026-06-15 |
| 45 | Claude Opus 4.8A | Anthropic | A | 917 | 940 | 99 | 2026-06-15 |
| 46 | Gemini 3.1 Flash Lite | Google Gemini | — | 957 | — | 99 | 2026-06-14 |
| 47 | Qwen 3.7 MaxA | OpenRouter | A | 973 | 2779 | — | 2026-06-15 |
| 48 | Claude Opus 4.6B | Anthropic | B | 973 | 1535 | 100 | 2026-06-15 |
| 49 | Gemini 2.5 ProA | Google Gemini | A | 1008 | 1089 | 0 | 2026-06-15 |
| 50 | gpt-4o-2024-05-13C | OpenAI | C | 1049 | — | 98 | 2026-06-14 |
| 51 | gpt-5-miniC | OpenAI | C | 1056 | 1850 | — | 2026-06-15 |
| 52 | gpt-4.1-2025-04-14 | OpenAI | — | 1072 | — | 99 | 2026-06-14 |
| 53 | gpt-5.5C | OpenAI | C | 1087 | 1762 | — | 2026-06-15 |
| 54 | Claude Opus 4.7B | Anthropic | B | 1105 | 3971 | 100 | 2026-06-15 |
| 55 | gpt-5.2B | OpenAI | B | 1118 | 1209 | — | 2026-06-15 |
| 56 | Qwen 3.6 PlusA | OpenRouter | A | 1169 | 1184 | — | 2026-06-15 |
| 57 | DeepSeek v3.2A | OpenRouter | A | 1202 | 1954 | — | 2026-06-15 |
| 58 | gpt-4o-2024-11-20C | OpenAI | C | 1326 | — | 99 | 2026-06-14 |
| 59 | gpt-3.5-turbo-1106 | OpenAI | — | 1328 | — | 97 | 2026-06-14 |
| 60 | Gemini Flash-Lite LatestC | Google Gemini | C | 1366 | — | 100 | 2026-06-14 |
| 61 | Claude Sonnet 4.5B | Anthropic | B | 1398 | 1794 | 99 | 2026-06-15 |
| 62 | Claude Sonnet 4.6A | Anthropic | A | 1648 | 2331 | 100 | 2026-06-15 |
| 63 | Nano Banana | Google Gemini | — | 1808 | — | 100 | 2026-06-14 |
| 64 | Claude Opus 4.5B | Anthropic | B | 1834 | 1922 | 100 | 2026-06-15 |
| 65 | Nano Banana 2 | Google Gemini | — | 1887 | — | 100 | 2026-06-14 |
| 66 | Claude Opus 4.1C | Anthropic | C | 1988 | 2300 | 100 | 2026-06-15 |
| 67 | gpt-3.5-turboC | OpenAI | C | 1995 | — | 95 | 2026-06-14 |
| 68 | gpt-3.5-turbo-16k | OpenAI | — | 2006 | — | 98 | 2026-06-14 |
| 69 | MiniMax M2.5A | OpenRouter | A | 2008 | 6695 | — | 2026-06-15 |
| 70 | Claude Opus 4C | Anthropic | C | 2009 | 2151 | 100 | 2026-06-15 |
| 71 | gpt-4o-2024-08-06C | OpenAI | C | 2016 | — | 99 | 2026-06-14 |
| 72 | gpt-4.1-nano-2025-04-14 | OpenAI | — | 2051 | — | 100 | 2026-06-14 |
| 73 | gpt-3.5-turbo-0125 | OpenAI | — | 2331 | — | 95 | 2026-06-14 |
| 74 | Gemini Robotics-ER 1.6 Preview | Google Gemini | — | 2764 | — | 99 | 2026-06-14 |
| 75 | Gemini 3 Flash PreviewC | Google Gemini | C | 2780 | — | 100 | 2026-06-14 |
| 76 | gpt-4o-search-previewC | OpenAI | C | 2930 | — | 95 | 2026-06-14 |
| 77 | gpt-4o-mini-search-previewC | OpenAI | C | 3388 | — | 98 | 2026-06-14 |
| 78 | gpt-5-search-apiC | OpenAI | C | 3559 | — | 99 | 2026-06-14 |
| 79 | gpt-4.1-mini-2025-04-14 | OpenAI | — | 3561 | — | 100 | 2026-06-14 |
| 80 | Gemini 3.5 FlashA | Google Gemini | A | 3938 | — | 88 | 2026-06-14 |
| 81 | gpt-4o-mini-2024-07-18C | OpenAI | C | 3960 | — | 98 | 2026-06-14 |
| 82 | Gemini Flash LatestB | Google Gemini | B | 4051 | — | 35 | 2026-06-14 |
| 83 | gpt-4o-mini-search-preview-2025-03-11 | OpenAI | — | 4627 | — | 72 | 2026-06-14 |
| 84 | gpt-4o-search-preview-2025-03-11 | OpenAI | — | 4883 | — | 99 | 2026-06-14 |
| 85 | gpt-5-search-api-2025-10-14 | OpenAI | — | 5351 | — | 99 | 2026-06-14 |
| 86 | gpt-4-0613 | OpenAI | — | 5810 | — | 91 | 2026-06-14 |
| 87 | Gemini 3.1 Pro Preview Custom ToolsC | Google Gemini | C | 6069 | — | 35 | 2026-06-14 |
| 88 | Gemini Pro LatestC | Google Gemini | C | 6574 | — | 51 | 2026-06-14 |
| 89 | gpt-4.1-miniC | OpenAI | C | 6924 | 30000 | 100 | 2026-06-15 |
| 90 | Gemini 3.1 Pro PreviewC | Google Gemini | C | 6937 | — | 40 | 2026-06-14 |
| 91 | gpt-4-turbo-2024-04-09C | OpenAI | C | 7386 | — | 99 | 2026-06-14 |
| 92 | gpt-4C | OpenAI | C | 7408 | — | 99 | 2026-06-14 |
| 93 | Nano Banana Pro | Google Gemini | — | 8045 | — | 25 | 2026-06-14 |
| 94 | gpt-4-turboC | OpenAI | C | 9151 | — | 99 | 2026-06-14 |
| 95 | Lyria 3 Clip Preview | Google Gemini | — | 9402 | — | 40 | 2026-06-14 |
| 96 | Gemma 4 31B ITC | Google Gemini | C | 11240 | — | 90 | 2026-06-14 |
| 97 | Gemma 4 26B A4B ITC | Google Gemini | C | 12943 | — | 95 | 2026-06-14 |
| 98 | Lyria 3 Pro Preview | Google Gemini | — | 21413 | — | 43 | 2026-05-13 |
98 of 98 models · click column headers to sort
Fast (< 500 ms)
Medium (500–1000 ms)
Slow (> 1000 ms)
Updated every 6 hours · P50 = median latency · P95 = tail latency