| 1 | — | Oo4-mini (high)闭源OpenAI | 69.8 | — | — | — |
| 2 | — | GGemini 2.5 Pro Preview (Mar' 25)闭源Google | 67.8 | — | — | — |
| 3 | — | Oo3闭源OpenAI | 66.8 | — | — | — |
| 4 | — | XGrok 3 mini Reasoning (high)闭源xAI | 66.7 | — | — | — |
| 5 | — | Oo3-mini (high)闭源OpenAI | 65.5 | — | — | — |
| 6 | — | Oo3-mini闭源OpenAI | 62.9 | — | — | — |
| 7 | — | Oo1闭源OpenAI | 61.9 | — | — | — |
| 8 | — | NLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)开源NVIDIA | 60.8 | — | — | — |
| 9 | — | GGemini 2.5 Flash Preview (Reasoning)闭源Google | 60.5 | — | — | — |
| 10 | — | DDeepSeek R1开源DeepSeek | 60.2 | — | — | — |
| 11 | — | AQwQ 32B开源Alibaba | 58.1 | — | — | — |
| 12 | — | AClaude 3.7 Sonnet (Extended Thinking)闭源Anthropic | 57.4 | — | — | — |
| 13 | — | Oo1-mini闭源OpenAI | 53.8 | — | — | — |
| 14 | — | DDeepSeek V3 0324 (Mar' 25)开源DeepSeek | 53.2 | — | — | — |
| 15 | — | OGPT-4.1 mini闭源OpenAI | 52.9 | — | — | — |
| 16 | — | OGPT-4.1闭源OpenAI | 52.6 | — | — | — |
| 17 | — | GGemini 2.0 Flash Thinking Experimental (Jan '25)闭源Google | 52.3 | — | — | — |
| 18 | — | DDeepSeek R1 Distill Qwen 32B开源DeepSeek | 51.9 | — | — | — |
| 19 | — | XGrok 3闭源xAI | 50.6 | — | — | — |
| 20 | — | MLlama 4 Maverick开源Meta | 50.5 | — | — | — |
| 21 | — | OGPT-4o (March 2025, chatgpt-4o-latest)闭源OpenAI | 50.3 | — | — | — |
| 22 | — | GGemini 2.0 Pro Experimental (Feb '25)闭源Google | 49.2 | — | — | — |
| 23 | — | DDeepSeek R1 Distill Qwen 14B开源DeepSeek | 49.0 | — | — | — |
| 24 | — | DDeepSeek R1 Distill Llama 70B开源DeepSeek | 48.2 | — | — | — |
| 25 | — | AClaude 3.7 Sonnet (Standard)闭源Anthropic | 48.2 | — | — | — |
| 26 | — | GGemini 2.0 Flash (Feb '25)闭源Google | 48.1 | — | — | — |
| 27 | — | RReka Flash 3开源Reka AI | 47.1 | — | — | — |
| 28 | — | GGemini 2.0 Flash (experimental)闭源Google | 46.4 | — | — | — |
| 29 | — | DDeepSeek V3 (Dec '24)开源DeepSeek | 45.6 | — | — | — |
| 30 | — | AQwen2.5 Max闭源Alibaba | 45.3 | — | — | — |
| 31 | — | GGemini 1.5 Pro (Sep '24)闭源Google | 44.6 | — | — | — |
| 32 | — | AClaude 3.5 Sonnet (Oct '24)闭源Anthropic | 44.4 | — | — | — |
| 33 | — | PSonar闭源Perplexity | 43.4 | — | — | — |
| 34 | — | MLlama 4 Scout开源Meta | 43.0 | — | — | — |
| 35 | — | PSonar Pro闭源Perplexity | 42.7 | — | — | — |
| 36 | — | AQwQ 32B-Preview开源Alibaba | 42.6 | — | — | — |
| 37 | — | OGPT-4o (Nov '24)闭源OpenAI | 41.5 | — | — | — |
| 38 | — | GGemini 2.0 Flash-Lite (Feb '25)闭源Google | 41.4 | — | — | — |
| 39 | — | MLlama 3.3 Instruct 70B开源Meta | 41.1 | — | — | — |
| 40 | — | OGPT-4.1 nano闭源OpenAI | 41.0 | — | — | — |
| 41 | — | OGPT-4o (May '24)闭源OpenAI | 40.9 | — | — | — |
| 42 | — | MLlama 3.1 Instruct 405B开源Meta | 40.5 | — | — | — |
| 43 | — | AQwen2.5 Instruct 72B开源Alibaba | 40.4 | — | — | — |
| 44 | — | MMiniMax-Text-01开源MiniMax | 40.2 | — | — | — |
| 45 | — | MPhi-4开源Microsoft Azure | 40.2 | — | — | — |
| 46 | — | CCommand A开源Cohere | 40.0 | — | — | — |
| 47 | — | ALlama 3.1 Tulu3 405B开源Allen Institute for AI | 39.9 | — | — | — |
| 48 | — | NLlama 3.3 Nemotron Super 49B v1开源NVIDIA | 39.3 | — | — | — |
| 49 | — | XGrok 2 (Dec '24)闭源xAI | 39.2 | — | — | — |
| 50 | — | GGemini 1.5 Flash (Sep '24)闭源Google | 39.0 | — | — | — |
| 51 | — | MMistral Large 2 (Nov '24)开源Mistral | 38.3 | — | — | — |
| 52 | — | GGemma 3 27B Instruct开源Google | 37.6 | — | — | — |
| 53 | — | XGrok Beta闭源xAI | 37.5 | — | — | — |
| 54 | — | MPixtral Large开源Mistral | 37.4 | — | — | — |
| 55 | — | AQwen2.5 Instruct 32B开源Alibaba | 37.4 | — | — | — |
| 56 | — | NLlama 3.1 Nemotron Instruct 70B开源NVIDIA | 37.3 | — | — | — |
| 57 | — | ANova Pro闭源Amazon | 37.1 | — | — | — |
| 58 | — | MMistral Large 2 (Jul '24)开源Mistral | 36.6 | — | — | — |
| 59 | — | AQwen2.5 Coder Instruct 32B开源Alibaba | 36.3 | — | — | — |
| 60 | — | OGPT-4o mini闭源OpenAI | 35.7 | — | — | — |
| 61 | — | MLlama 3.1 Instruct 70B开源Meta | 35.4 | — | — | — |
| 62 | — | MMistral Small 3.1开源Mistral | 35.3 | — | — | — |
| 63 | — | MMistral Small 3开源Mistral | 35.3 | — | — | — |
| 64 | — | AClaude 3 Opus闭源Anthropic | 35.1 | — | — | — |
| 65 | — | AClaude 3.5 Haiku闭源Anthropic | 34.7 | — | — | — |
| 66 | — | DDeepSeek R1 Distill Llama 8B开源DeepSeek | 34.0 | — | — | — |
| 67 | — | GGemma 3 12B Instruct开源Google | 33.8 | — | — | — |
| 68 | — | GGemini 1.5 Pro (May '24)闭源Google | 33.7 | — | — | — |
| 69 | — | AQwen Turbo闭源Alibaba | 33.6 | — | — | — |
| 70 | — | MLlama 3.2 Instruct 90B (Vision)开源Meta | 33.4 | — | — | — |
| 71 | — | AQwen2 Instruct 72B开源Alibaba | 32.6 | — | — | — |
| 72 | — | ANova Lite闭源Amazon | 32.5 | — | — | — |
| 73 | — | GGemini 1.5 Flash-8B闭源Google | 30.8 | — | — | — |
| 74 | — | NDeepHermes 3 - Mistral 24B Preview开源Nous Research | 30.0 | — | — | — |
| 75 | — | AJamba 1.5 Large开源AI21 Labs | 29.3 | — | — | — |
| 76 | — | NHermes 3 - Llama-3.1 70B开源Nous Research | 29.1 | — | — | — |
| 77 | — | AJamba 1.6 Large开源AI21 Labs | 28.8 | — | — | — |
| 78 | — | GGemini 1.5 Flash (May '24)闭源Google | 28.5 | — | — | — |
| 79 | — | ANova Micro闭源Amazon | 28.3 | — | — | — |
| 80 | — | 0Yi-Large闭源01.AI | 27.9 | — | — | — |
| 81 | — | AClaude 3 Sonnet闭源Anthropic | 27.8 | — | — | — |
| 82 | — | MCodestral (Jan '25)闭源Mistral | 27.6 | — | — | — |
| 83 | — | MLlama 3 Instruct 70B开源Meta | 27.5 | — | — | — |
| 84 | — | MMistral Small (Sep '24)开源Mistral | 27.4 | — | — | — |
| 85 | — | MPhi-4 Multimodal Instruct开源Microsoft Azure | 26.9 | — | — | — |
| 86 | — | AQwen2.5 Coder Instruct 7B开源Alibaba | 26.7 | — | — | — |
| 87 | — | MMistral Large (Feb '24)闭源Mistral | 26.4 | — | — | — |
| 88 | — | MMixtral 8x22B Instruct开源Mistral | 26.2 | — | — | — |
| 89 | — | MPhi-4 Mini Instruct开源Microsoft Azure | 26.0 | — | — | — |
| 90 | — | MPhi-3 Medium Instruct 14B开源Microsoft Azure | 24.5 | — | — | — |
| 91 | — | GGemma 3 4B Instruct开源Google | 24.5 | — | — | — |
| 92 | — | AClaude 2.1闭源Anthropic | 24.1 | — | — | — |
| 93 | — | MLlama 3.1 Instruct 8B开源Meta | 23.7 | — | — | — |
| 94 | — | MPixtral 12B (2409)开源Mistral | 23.4 | — | — | — |
| 95 | — | MMistral Small (Feb '24)闭源Mistral | 22.9 | — | — | — |
| 96 | — | MMistral Medium闭源Mistral | 22.8 | — | — | — |
| 97 | — | MMinistral 8B开源Mistral | 22.3 | — | — | — |
| 98 | — | GGemma 2 9B开源Google | 22.2 | — | — | — |
| 99 | — | MPhi-3 Mini Instruct 3.8B开源Microsoft Azure | 22.1 | — | — | — |
| 100 | — | LLFM 40B闭源Liquid AI | 21.7 | — | — | — |
| 101 | — | CCommand-R+ (Aug '24)开源Cohere | 21.5 | — | — | — |
| 102 | — | MLlama 3 Instruct 8B开源Meta | 21.5 | — | — | — |
| 103 | — | GGemini 1.0 Pro闭源Google | 20.6 | — | — | — |
| 104 | — | MCodestral (May '24)开源Mistral | 20.4 | — | — | — |
| 105 | — | CAya Expanse 32B开源Cohere | 20.1 | — | — | — |
| 106 | — | MLlama 2 Chat 13B开源Meta | 19.9 | — | — | — |
| 107 | — | CCommand-R+ (Apr '24)开源Cohere | 19.9 | — | — | — |
| 108 | — | DDBRX Instruct开源Databricks | 19.8 | — | — | — |
| 109 | — | MMinistral 3B闭源Mistral | 19.6 | — | — | — |
| 110 | — | MMistral NeMo开源Mistral | 19.6 | — | — | — |
| 111 | — | MLlama 3.2 Instruct 3B开源Meta | 19.5 | — | — | — |
| 112 | — | DDeepSeek R1 Distill Qwen 1.5B开源DeepSeek | 19.2 | — | — | — |
| 113 | — | AJamba 1.5 Mini开源AI21 Labs | 18.4 | — | — | — |
| 114 | — | AJamba 1.6 Mini开源AI21 Labs | 17.7 | — | — | — |
| 115 | — | MMixtral 8x7B Instruct开源Mistral | 17.0 | — | — | — |
| 116 | — | NDeepHermes 3 - Llama-3.1 8B Preview开源Nous Research | 16.2 | — | — | — |
| 117 | — | CAya Expanse 8B开源Cohere | 16.0 | — | — | — |
| 118 | — | CCommand-R (Aug '24)开源Cohere | 14.8 | — | — | — |
| 119 | — | CCommand-R (Mar '24)开源Cohere | 14.7 | — | — | — |
| 120 | — | MCodestral-Mamba开源Mistral | 13.9 | — | — | — |
| 121 | — | MMistral 7B Instruct开源Mistral | 10.1 | — | — | — |
| 122 | — | MLlama 3.2 Instruct 1B开源Meta | 9.7 | — | — | — |
| 123 | — | MLlama 2 Chat 7B开源Meta | 8.3 | — | — | — |
