| 1 | — | OGPT-5 (high)闭源OpenAI | 68.5 | $3.4 | — | — |
| 2 | 新 | OGPT-5 Codex (high)闭源OpenAI | 67.6 | $3.4 | — | — |
| 3 | ▼ 1 | OGPT-5 (medium)闭源OpenAI | 66.4 | $3.4 | — | — |
| 4 | ▲ 1 | Oo3闭源OpenAI | 65.5 | $3.5 | — | — |
| 5 | ▼ 2 | XGrok 4闭源xAI | 65.3 | $6.0 | — | — |
| 6 | ▼ 2 | Oo3-pro闭源OpenAI | 65.3 | $35 | — | — |
| 7 | ▼ 1 | OGPT-5 mini (high)闭源OpenAI | 64.3 | $0.69 | — | — |
| 8 | ▼ 1 | OGPT-5 (low)闭源OpenAI | 61.8 | $3.4 | — | — |
| 9 | 新 | AClaude 4.5 Sonnet (Reasoning)闭源Anthropic | 61.3 | $6.0 | — | — |
| 10 | ▼ 2 | OGPT-5 mini (medium)闭源OpenAI | 60.8 | $0.69 | — | — |
| 11 | ▼ 2 | XGrok 4 Fast (Reasoning)闭源xAI | 60.3 | $0.28 | — | — |
| 12 | ▼ 2 | GGemini 2.5 Pro闭源Google | 59.6 | $3.4 | — | — |
| 13 | ▼ 1 | Oo4-mini (high)闭源OpenAI | 59.6 | $1.9 | — | — |
| 14 | 新 | AClaude 4.1 Opus (Reasoning)闭源Anthropic | 59.3 | $30 | — | — |
| 15 | ▼ 2 | Ogpt-oss-120B (high)开源OpenAI | 58.0 | $0.26 | — | — |
| 16 | ▼ 2 | XGrok 3 mini Reasoning (high)闭源xAI | 57.1 | $0.35 | — | — |
| 17 | 新 | AClaude 4 Sonnet (Reasoning)闭源Anthropic | 56.5 | $6.0 | — | — |
| 18 | 新 | GGemini 2.5 Flash Preview (Sep '25) (Reasoning)闭源Google | 54.4 | $0.85 | — | — |
| 19 | 新 | AClaude 4 Opus (Reasoning)闭源Anthropic | 54.2 | $30 | — | — |
| 20 | ▼ 3 | GGemini 2.5 Pro Preview (Mar' 25)闭源Google | 54.1 | $3.4 | — | — |
| 21 | ▼ 3 | GGemini 2.5 Pro Preview (May' 25)闭源Google | 53.2 | $3.4 | — | — |
| 22 | ▼ 3 | GGemini 2.5 Flash (Reasoning)闭源Google | 51.2 | $0.85 | — | — |
| 23 | — | OGPT-5 nano (high)闭源OpenAI | 51.0 | $0.14 | — | — |
| 24 | ▼ 4 | Oo3-mini (high)闭源OpenAI | 50.8 | $1.9 | — | — |
| 25 | 新 | AClaude 3.7 Sonnet (Reasoning)闭源Anthropic | 49.9 | $6.0 | — | — |
| 26 | — | OGPT-5 nano (medium)闭源OpenAI | 49.3 | $0.14 | — | — |
| 27 | ▼ 5 | XGrok Code Fast 1闭源xAI | 48.7 | $0.53 | — | — |
| 28 | 新 | AClaude 4.5 Sonnet (Non-reasoning)闭源Anthropic | 48.5 | $6.0 | — | — |
| 29 | ▼ 5 | Oo3-mini闭源OpenAI | 48.1 | $1.9 | — | — |
| 30 | ▼ 5 | Oo1-pro闭源OpenAI | 48.0 | $263 | — | — |
| 31 | 新 | GGemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)闭源Google | 47.9 | $0.18 | — | — |
| 32 | ▼ 5 | Oo1闭源OpenAI | 47.2 | $26 | — | — |
| 33 | 新 | GGemini 2.5 Flash Preview (Sep '25) (Non-reasoning)闭源Google | 46.7 | $0.85 | — | — |
| 34 | ▼ 6 | PSonar Reasoning Pro闭源Perplexity | 46.3 | $0.000 | — | — |
| 35 | ▼ 6 | GGemini 2.5 Flash Preview (Reasoning)闭源Google | 45.8 | $0.000 | — | — |
| 36 | ▼ 6 | NLlama Nemotron Super 49B v1.5 (Reasoning)开源NVIDIA | 45.2 | $0.18 | — | — |
| 37 | ▼ 6 | Oo1-preview闭源OpenAI | 44.9 | $29 | — | — |
| 38 | 新 | AClaude 4.1 Opus (Non-reasoning)闭源Anthropic | 44.6 | $30 | — | — |
| 39 | 新 | AClaude 4 Sonnet (Non-reasoning)闭源Anthropic | 44.4 | $6.0 | — | — |
| 40 | ▼ 5 | OGPT-5 (minimal)闭源OpenAI | 43.5 | $3.4 | — | — |
| 41 | ▼ 5 | OGPT-4.1闭源OpenAI | 43.4 | $3.5 | — | — |
| 42 | ▼ 10 | Ogpt-oss-20B (high)开源OpenAI | 43.3 | $0.088 | — | — |
| 43 | ▼ 6 | OGPT-4.1 mini闭源OpenAI | 42.5 | $0.70 | — | — |
| 44 | 新 | AClaude 4 Opus (Non-reasoning)闭源Anthropic | 42.3 | $30 | — | — |
| 45 | ▼ 6 | OGPT-5 (ChatGPT)闭源OpenAI | 41.8 | $3.4 | — | — |
| 46 | 新 | GGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)闭源Google | 41.6 | $0.18 | — | — |
| 47 | ▼ 7 | OGPT-5 mini (minimal)闭源OpenAI | 41.6 | $0.69 | — | — |
| 48 | ▼ 7 | XGrok 3 Reasoning Beta闭源xAI | 41.4 | $0.000 | — | — |
| 49 | 新 | GGemini 2.5 Flash (Non-reasoning)闭源Google | 40.4 | $0.85 | — | — |
| 50 | ▼ 7 | GGemini 2.5 Flash-Lite (Reasoning)闭源Google | 40.1 | $0.18 | — | — |
| 51 | ▼ 7 | Oo1-mini闭源OpenAI | 39.2 | $1.9 | — | — |
| 52 | 新 | XGrok 4 Fast (Non-reasoning)闭源xAI | 38.6 | $0.28 | — | — |
| 53 | ▼ 7 | NLlama 3.1 Nemotron Ultra 253B v1 (Reasoning)开源NVIDIA | 38.5 | $0.90 | — | — |
| 54 | ▼ 7 | OGPT-4.5 (Preview)闭源OpenAI | 38.4 | $0.000 | — | — |
| 55 | ▼ 6 | GGemini 2.0 Flash Thinking Experimental (Jan '25)闭源Google | 37.7 | $0.000 | — | — |
| 56 | ▼ 6 | NNVIDIA Nemotron Nano 9B V2 (Reasoning)开源NVIDIA | 36.9 | $0.070 | — | — |
| 57 | 新 | NNVIDIA Nemotron Nano 9B V2 (Non-reasoning)开源NVIDIA | 36.1 | $0.070 | — | — |
| 58 | ▼ 7 | XGrok 3闭源xAI | 36.0 | $6.0 | — | — |
| 59 | ▼ 7 | MLlama 4 Maverick开源Meta | 35.8 | $0.39 | — | — |
| 60 | ▼ 7 | OGPT-4o (March 2025, chatgpt-4o-latest)闭源OpenAI | 35.6 | $7.5 | — | — |
| 61 | ▼ 7 | NLlama 3.3 Nemotron Super 49B v1 (Reasoning)开源NVIDIA | 35.5 | $0.000 | — | — |
| 62 | ▼ 7 | GGemini 2.0 Pro Experimental (Feb '25)闭源Google | 34.6 | $0.000 | — | — |
| 63 | ▼ 7 | PSonar Reasoning闭源Perplexity | 34.2 | $2.0 | — | — |
| 64 | 新 | GGemini 2.5 Flash Preview (Non-reasoning)闭源Google | 34.1 | $0.000 | — | — |
| 65 | ▼ 7 | GGemini 2.0 Flash (Feb '25)闭源Google | 33.6 | $0.18 | — | — |
| 66 | 新 | AClaude 3.7 Sonnet (Non-reasoning)闭源Anthropic | 33.5 | $6.0 | — | — |
| 67 | ▼ 6 | ANova Premier闭源Amazon | 32.3 | $5.0 | — | — |
| 68 | ▼ 8 | GGemini 2.0 Flash (experimental)闭源Google | 31.8 | $0.000 | — | — |
| 69 | 新 | GGemini 2.5 Flash-Lite (Non-reasoning)闭源Google | 30.1 | $0.18 | — | — |
| 70 | ▼ 7 | GGemini 1.5 Pro (Sep '24)闭源Google | 30.0 | $0.000 | — | — |
| 71 | ▼ 7 | AClaude 3.5 Sonnet (Oct '24)闭源Anthropic | 29.9 | $6.0 | — | — |
| 72 | ▼ 7 | OGPT-5 nano (minimal)闭源OpenAI | 29.1 | $0.14 | — | — |
| 73 | ▼ 7 | PSonar闭源Perplexity | 28.8 | $1.0 | — | — |
| 74 | ▼ 7 | PSonar Pro闭源Perplexity | 28.2 | $6.0 | — | — |
| 75 | ▼ 7 | MLlama 4 Scout开源Meta | 28.1 | $0.26 | — | — |
| 76 | ▼ 7 | MLlama 3.3 Instruct 70B开源Meta | 27.9 | $0.57 | — | — |
| 77 | ▼ 7 | OGPT-4.1 nano闭源OpenAI | 27.3 | $0.18 | — | — |
| 78 | ▼ 7 | OGPT-4o (Nov '24)闭源OpenAI | 27.0 | $4.4 | — | — |
| 79 | ▼ 7 | GGemini 2.0 Flash-Lite (Feb '25)闭源Google | 26.8 | $0.13 | — | — |
| 80 | 新 | NLlama Nemotron Super 49B v1.5 (Non-reasoning)开源NVIDIA | 26.6 | $0.18 | — | — |
| 81 | ▼ 7 | OGPT-4o (May '24)闭源OpenAI | 26.3 | $7.5 | — | — |
| 82 | ▼ 7 | GGemini 2.0 Flash-Lite (Preview)闭源Google | 26.3 | $0.13 | — | — |
| 83 | ▼ 7 | NLlama 3.1 Nemotron Nano 4B v1.1 (Reasoning)开源NVIDIA | 26.1 | $0.000 | — | — |
| 84 | ▼ 7 | OGPT-4o (Aug '24)闭源OpenAI | 26.0 | $4.4 | — | — |
| 85 | 新 | NLlama 3.3 Nemotron Super 49B v1 (Non-reasoning)开源NVIDIA | 25.9 | $0.000 | — | — |
| 86 | ▼ 7 | MLlama 3.1 Instruct 405B开源Meta | 25.7 | $3.5 | — | — |
| 87 | ▼ 7 | ANova Pro闭源Amazon | 25.5 | $1.4 | — | — |
| 88 | ▼ 7 | AClaude 3.5 Sonnet (June '24)闭源Anthropic | 25.4 | $6.0 | — | — |
| 89 | ▼ 7 | OGPT-4o (ChatGPT)闭源OpenAI | 25.3 | $7.5 | — | — |
| 90 | ▼ 7 | XGrok 2 (Dec '24)开源xAI | 24.7 | $0.000 | — | — |
| 91 | ▼ 7 | MPhi-4开源Microsoft Azure | 24.6 | $0.22 | — | — |
| 92 | ▼ 7 | GGemini 1.5 Flash (Sep '24)闭源Google | 24.4 | $0.000 | — | — |
| 93 | ▼ 7 | OGPT-4 Turbo闭源OpenAI | 24.2 | $15 | — | — |
| 94 | ▼ 7 | NLlama 3.1 Nemotron Instruct 70B开源NVIDIA | 23.6 | $0.60 | — | — |
| 95 | ▼ 7 | XGrok Beta闭源xAI | 23.0 | $0.000 | — | — |
| 96 | ▼ 7 | MLlama 3.1 Instruct 70B开源Meta | 22.6 | $0.76 | — | — |
| 97 | ▼ 7 | GGemma 3 27B Instruct开源Google | 21.6 | $0.000 | — | — |
| 98 | ▼ 7 | OGPT-4闭源OpenAI | 21.5 | $38 | — | — |
| 99 | ▼ 7 | ANova Lite闭源Amazon | 21.4 | $0.11 | — | — |
| 100 | ▼ 7 | OGPT-4o mini闭源OpenAI | 21.2 | $0.26 | — | — |
| 101 | ▼ 6 | AClaude 3 Opus闭源Anthropic | 20.6 | $30 | — | — |
| 102 | ▼ 8 | GGemma 3 12B Instruct开源Google | 20.4 | $0.24 | — | — |
| 103 | ▼ 7 | AClaude 3.5 Haiku闭源Anthropic | 20.2 | $1.6 | — | — |
| 104 | ▼ 7 | GGemini 2.0 Flash Thinking Experimental (Dec '24)闭源Google | 20.2 | $0.000 | — | — |
| 105 | ▼ 7 | GGemini 1.5 Pro (May '24)闭源Google | 19.2 | $0.000 | — | — |
| 106 | ▼ 7 | PR1 1776开源Perplexity | 19.1 | $3.5 | — | — |
| 107 | ▼ 7 | MLlama 3.2 Instruct 90B (Vision)开源Meta | 18.9 | $0.72 | — | — |
| 108 | ▼ 7 | XGrok-1开源xAI | 18.2 | $0.000 | — | — |
| 109 | ▼ 7 | ANova Micro闭源Amazon | 17.3 | $0.061 | — | — |
| 110 | ▼ 7 | GGemma 2 27B开源Google | 17.2 | $0.000 | — | — |
| 111 | ▼ 7 | MLlama 3.1 Instruct 8B开源Meta | 16.9 | $0.10 | — | — |
| 112 | ▼ 7 | GGemini 1.5 Flash-8B闭源Google | 16.3 | $0.000 | — | — |
| 113 | ▼ 7 | MPhi-4 Mini Instruct开源Microsoft Azure | 15.7 | $0.000 | — | — |
| 114 | ▼ 6 | MLlama 3.2 Instruct 11B (Vision)开源Meta | 15.5 | $0.16 | — | — |
| 115 | ▼ 8 | GGemma 3n E4B Instruct开源Google | 15.5 | $0.025 | — | — |
| 116 | 新 | IGranite 3.3 8B (Non-reasoning)开源IBM | 15.2 | $0.085 | — | — |
| 117 | ▼ 7 | GGemma 3 4B Instruct开源Google | 14.7 | $0.050 | — | — |
| 118 | ▼ 7 | GGemini 1.5 Flash (May '24)闭源Google | 14.0 | $0.000 | — | — |
| 119 | ▼ 7 | AClaude 3 Sonnet闭源Anthropic | 13.3 | $6.0 | — | — |
| 120 | ▼ 7 | MPhi-3 Mini Instruct 3.8B开源Microsoft Azure | 13.1 | $0.23 | — | — |
| 121 | ▼ 7 | MLlama 3 Instruct 70B开源Meta | 13.0 | $0.75 | — | — |
| 122 | ▼ 7 | GGemini 1.0 Ultra闭源Google | 12.8 | $0.000 | — | — |
| 123 | ▼ 7 | GGemma 3n E4B Instruct Preview (May '25)开源Google | 12.5 | $0.000 | — | — |
| 124 | ▼ 7 | MPhi-4 Multimodal Instruct开源Microsoft Azure | 12.4 | $0.000 | — | — |
| 125 | 新 | LLFM2 2.6B开源Liquid AI | 11.8 | $0.000 | — | — |
| 126 | ▼ 8 | MLlama 2 Chat 7B开源Meta | 11.3 | $0.10 | — | — |
| 127 | ▼ 8 | MLlama 3.2 Instruct 3B开源Meta | 11.2 | $0.048 | — | — |
| 128 | ▼ 8 | MPhi-3 Medium Instruct 14B开源Microsoft Azure | 10.1 | $0.30 | — | — |
| 129 | ▼ 8 | AClaude 2.1闭源Anthropic | 9.7 | $0.000 | — | — |
| 130 | ▼ 8 | AClaude 3 Haiku闭源Anthropic | 9.6 | $0.50 | — | — |
| 131 | ▼ 8 | AClaude 2.0闭源Anthropic | 8.6 | $0.000 | — | — |
| 132 | ▼ 8 | OGPT-3.5 Turbo闭源OpenAI | 8.3 | $0.75 | — | — |
| 133 | ▼ 8 | LLFM2 1.2B开源Liquid AI | 8.0 | $0.000 | — | — |
| 134 | ▼ 8 | GGemma 3n E2B Instruct开源Google | 8.0 | $0.000 | — | — |
| 135 | ▼ 8 | GGemma 2 9B开源Google | 7.8 | $0.20 | — | — |
| 136 | ▼ 8 | SArctic Instruct开源Snowflake | 7.6 | $0.000 | — | — |
| 137 | ▼ 8 | LLFM 40B闭源Liquid AI | 7.3 | $0.000 | — | — |
| 138 | ▼ 8 | MLlama 3.2 Instruct 1B开源Meta | 7.1 | $0.053 | — | — |
| 139 | ▼ 8 | MLlama 3 Instruct 8B开源Meta | 7.0 | $0.055 | — | — |
| 140 | ▼ 8 | GPALM-2闭源Google | 6.6 | $0.000 | — | — |
| 141 | ▼ 8 | GGemini 1.0 Pro闭源Google | 6.2 | $0.75 | — | — |
| 142 | ▼ 8 | GGemma 3 1B Instruct开源Google | 6.1 | $0.000 | — | — |
| 143 | ▼ 8 | MLlama 2 Chat 70B开源Meta | 5.6 | $0.000 | — | — |
| 144 | ▼ 8 | MLlama 2 Chat 13B开源Meta | 5.5 | $0.000 | — | — |
| 145 | ▼ 8 | DDBRX Instruct开源Databricks | 5.3 | $0.000 | — | — |
| 146 | ▼ 8 | AClaude Instant闭源Anthropic | 1.0 | $0.000 | — | — |
| 146 | ▼ 8 | MLlama 65B开源Meta | 1.0 | $0.000 | — | — |