2026-10-10最新大模型综合评分排行榜
数据更新于 2026-10-10
榜单点评:这张榜单像是 Anthropic 的主场:Claude 占前五席中的四席、前二十席中的六席;OpenAI 则有五席进前二十,但只有一个进入前三。更反常的是,版本号没能预测名次:Gemini 3.7 Flash 排在 3.8 Flash 前面,Claude Opus 4.7 也高于 4.8。Google 的两款 Gemini 都在前二十,却没有进入前十。新版、旧版和厂商席位,几条线索摆在一起,比单看榜首更耐琢磨。
| 排名 | 模型 | 评分 | 输入价格 | 输出价格 | OR 输入价 | OR 输出价 | 速度 | 对比 |
|---|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 Anthropic | 86.5 | $4/M | $20/M | $4/M | $20/M | 92 tokens/秒 | |
| 2 | Claude Fable 5.1 Anthropic | 84.77 | $10/M | $50/M | $10/M | $50/M | 69 tokens/秒 | |
| 3 | GPT-6.1 Sol OpenAI | 81.69 | — | — | — | — | — | |
| 4 | Claude Sonnet 5.5 Anthropic | 80.01 | — | — | — | — | — | |
| 5 | Claude Opus 5 Anthropic | 79.84 | $5/M | $25/M | $5/M | $25/M | 52 tokens/秒 | |
| 6 | GPT-6 Astra OpenAI | 79.27 | $10/M | $50/M | $10/M | $50/M | 51 tokens/秒 | |
| 7 | Muse Spark 1.3 Meta | 79.03 | — | — | — | — | — | |
| 8 | GPT-5.6 Sol OpenAI | 76.41 | $4/M | $20/M | $2/M | $10/M | 80 tokens/秒 | |
| 9 | MiMo-V2.6-Pro 小米 | 74.38 | — | — | — | — | — | |
| 10 | Kimi K3 月之暗面 | 73.84 | $3/M | $15/M | $0.67/M | $14/M | 34 tokens/秒 | |
| 11 | GPT-6 Sol OpenAI | 72.84 | $2/M | $10/M | $2/M | $10/M | — | |
| 12 | Gemini 3.7 Flash 谷歌 | 72.39 | $0.75/M | $3.75/M | $0.75/M | $3.75/M | 296 tokens/秒 | |
| 13 | Claude Opus 4.7 Anthropic | 72.34 | $5/M | $25/M | $5/M | $25/M | 47 tokens/秒 | |
| 14 | GPT-5.5 OpenAI | 72.18 | $5/M | $30/M | $5/M | $30/M | 90 tokens/秒 | |
| 15 | Qwen3.8 Max 阿里通义 | 71.16 | — | — | $0.78/M | $3.9/M | 39 tokens/秒 | |
| 16 | Gemini 3.8 Flash 谷歌 | 70.91 | $0.75/M | $3.75/M | $0.75/M | $3.75/M | 236 tokens/秒 | |
| 17 | Claude Opus 4.8 Anthropic | 70.46 | $5/M | $25/M | $5/M | $25/M | 59 tokens/秒 | |
| 18 | MiMo-V2.6-Flash 小米 | 68.02 | — | — | — | — | — | |
| 19 | Grok 4.6 xAI | 67.78 | $2/M | $6/M | $2/M | $6/M | 64 tokens/秒 | |
| 20 | GLM-5.3 智谱 AI | 67.62 | 免费 | 免费 | $1.4/M | $4.4/M | 68 tokens/秒 | |
| 21 | Claude Opus 4.6 Anthropic | 67.53 | $5/M | $25/M | $5/M | $25/M | 37 tokens/秒 | |
| 22 | DeepSeek V4.1 Flash 深度求索 | 67.07 | — | — | — | — | — | |
| 23 | Grok 4.7 xAI | 66.56 | — | — | — | — | — | |
| 24 | GPT-5.6 Terra OpenAI | 66.53 | $2/M | $12/M | $2/M | $12/M | — | |
| 25 | Muse Spark 1.2 Meta | 66.29 | — | — | — | — | — | |
| 26 | Muse Spark 1.1 Meta | 66.04 | — | — | — | — | — | |
| 27 | Gemini 3.6 Flash 谷歌 | 63.71 | $0.75/M | $3.75/M | $0.75/M | $3.75/M | 184 tokens/秒 | |
| 28 | Gemini 3.5 Flash 谷歌 | 63.21 | $1.5/M | $9/M | $1.5/M | $9/M | 200 tokens/秒 | |
| 29 | Qwen3.7 Max 阿里通义 | 62.83 | — | — | — | — | — | |
| 30 | Grok 4.5 xAI | 62.25 | $2/M | $6/M | $2/M | $6/M | 57 tokens/秒 | |
| 31 | Gemini 3.1 Pro 谷歌 | 61.95 | $2/M | $12/M | — | — | 113 tokens/秒 | |
| 32 | GLM-5.3-Flash 智谱 AI | 61.53 | — | — | — | — | — | |
| 33 | Claude Opus 4.5 Anthropic | 61.35 | $5/M | $25/M | $5/M | $25/M | 45 tokens/秒 | |
| 34 | GLM-5.2 智谱 AI | 60.68 | $1.4/M | $4.4/M | $0.0224/M | $16/M | 84 tokens/秒 | |
| 35 | Claude Sonnet 5 Anthropic | 60.58 | $2/M | $10/M | $2/M | $10/M | 76 tokens/秒 | |
| 36 | DeepSeek V4 Pro 0813 深度求索 | 58.73 | $1.32/M | $3.96/M | $0.55/M | $5/M | 180 tokens/秒 | |
| 37 | GPT-6 Luna OpenAI | 58.53 | $0.1/M | $0.5/M | $0.1/M | $0.5/M | — | |
| 38 | GPT-5.4 OpenAI | 58.23 | $2.5/M | $15/M | $2.5/M | $15/M | 96 tokens/秒 | |
| 39 | GPT-5.2-Codex OpenAI | 55.3 | — | — | — | — | — | |
| 40 | Claude Sonnet 4.6 Anthropic | 55.15 | $3/M | $15/M | $3/M | $15/M | 43 tokens/秒 | |
| 41 | GPT-5.6 Luna OpenAI | 54.23 | $0.2/M | $1.2/M | $0.2/M | $1.2/M | — | |
| 42 | Qwen3.8-27B 阿里通义 | 53.34 | — | — | — | — | — | |
| 43 | GLM-5.1 智谱 AI | 53.2 | — | — | — | — | — | |
| 44 | MiMo-V2.5-Pro 小米 | 50.09 | — | — | — | — | — | |
| 45 | GPT-5.1 OpenAI | 49.45 | $1.25/M | $10/M | $1.25/M | $10/M | 101 tokens/秒 | |
| 46 | Qwen3.6 Plus 阿里通义 | 49.16 | — | — | $0.325/M | $1.95/M | 56 tokens/秒 | |
| 47 | DeepSeek V4 Flash 0731 深度求索 | 48.62 | — | — | — | — | — | |
| 48 | GLM-5 智谱 AI | 47.28 | — | — | — | — | — | |
| 49 | Qwen3.7 Plus 阿里通义 | 46.33 | — | — | $0.32/M | $1.28/M | 56 tokens/秒 | |
| 50 | GPT-5.2 OpenAI | 46.24 | $1.75/M | $14/M | $1.75/M | $14/M | 68 tokens/秒 | |
| 51 | MiniMax M3 MiniMax | 45.81 | $0.3/M | $1.2/M | $0.3/M | $1.2/M | 99 tokens/秒 | |
| 52 | Gemini 3.5 Flash-Lite 谷歌 | 45.55 | — | — | — | — | — | |
| 53 | Kimi K2.7 Code 月之暗面 | 44.9 | — | — | — | — | — | |
| 54 | Kimi K2.5 月之暗面 | 42.71 | — | — | — | — | — | |
| 55 | Qwen3.5-27B 阿里通义 | 40.83 | — | — | — | — | — | |
| 56 | Claude Sonnet 4.5 Anthropic | 40.16 | $3/M | $15/M | $3/M | $15/M | — | |
| 57 | MiMo-V2.5 小米 | 40.08 | — | — | — | — | — | |
| 58 | GLM-4.7 智谱 AI | 38.62 | — | — | — | — | — | |
| 59 | Grok 4.3 xAI | 38.28 | — | — | — | — | — | |
| 60 | GPT-5.4 mini OpenAI | 37.95 | — | — | — | — | — | |
| 61 | GPT-5.1-Codex OpenAI | 37.22 | — | — | — | — | — | |
| 62 | o3 OpenAI | 36.95 | $2/M | $8/M | $2/M | $8/M | 113 tokens/秒 | |
| 63 | MiniMax M2.7 MiniMax | 35.33 | — | — | — | — | — | |
| 64↑1 | Gemini 3.1 Flash-Lite 谷歌 | 33 | — | — | — | — | — | |
| 65↓1 | GLM-4.6 智谱 AI | 32.88 | — | — | — | — | — | |
| 66 | Grok 4 xAI | 32.73 | — | — | — | — | — | |
| 67 | MiniMax M2.5 MiniMax | 31.18 | — | — | — | — | — | |
| 68 | GPT-5 mini OpenAI | 31.07 | $0.25/M | $2/M | $0.25/M | $2/M | 86 tokens/秒 | |
| 69 | DeepSeek V3.2 深度求索 | 30.8 | — | — | — | — | — | |
| 70 | Gemma 4 26B A4B 谷歌 | 30.34 | — | — | — | — | — | |
| 71 | GPT-5.4 nano OpenAI | 29.24 | — | — | — | — | — | |
| 72 | Qwen3.5-35B-A3B 阿里通义 | 27.04 | — | — | — | — | — | |
| 73 | Kimi K2 月之暗面 | 26.8 | — | — | — | — | — | |
| 74 | Grok 4.1 Fast xAI | 24.84 | — | — | — | — | — | |
| 75 | Claude Haiku 4.5 Anthropic | 22.8 | — | — | — | — | — | |
| 76 | GPT-4.1 OpenAI | 22.31 | — | — | — | — | — | |
| 77 | Grok Code Fast 1 xAI | 20.53 | — | — | — | — | — | |
| 78 | GLM-4.7-Flash 智谱 AI | 19.9 | — | — | — | — | — | |
| 79 | GPT-5 nano OpenAI | 19.13 | — | — | — | — | — | |
| 80 | Gemini 2.5 Flash 谷歌 | 18.77 | — | — | — | — | — | |
| 81 | GLM-4.5-Air 智谱 AI | 18.28 | — | — | — | — | — | |
| 82 | DeepSeek V3 0324 深度求索 | 17.03 | — | — | — | — | — | |
| 83 | GPT-OSS 120B OpenAI | 15.89 | 免费 | 免费 | $0.037/M | $0.17/M | 174 tokens/秒 | |
| 84 | GPT-4.1 mini OpenAI | 15.54 | — | — | — | — | — | |
| 85 | GPT-OSS 20B OpenAI | 10.78 | — | — | — | — | — | |
| 86 | GPT-4.1 nano OpenAI | 7.23 | — | — | — | — | — |