2026-10-29最新大模型知识能力排行榜
数据更新于 2026-10-29
榜单点评:Claude Opus 5.5 以 86.57 分登顶,领先 GPT-6.1 Sol 2.76 分;但榜单的主角不止冠军:Anthropic 前十占了 5 席,OpenAI 也有 4 席,几乎把前十变成两家轮流排座。第二、三名 GPT-6.1 Sol 与 GPT-6 Astra 仅差 0.07 分,名次就像贴着走。Google 的 Gemini 4 Argon 排第六,是前十里唯一的 Google 模型;小米、Moonshot、阶跃星辰和 Qwen 则各有一个模型进入前二十。冠军归 Claude,前十却是 Anthropic 与 OpenAI 的双人局。
| 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 | 86.57 | $4/M | $20/M | $4/M | $20/M | 97 tokens/秒 | |
| 2 | GPT-6.1 Sol OpenAI | 83.81 | $2/M | $10/M | $2/M | $10/M | — | |
| 3 | GPT-6 Astra OpenAI | 83.74 | $10/M | $50/M | $10/M | $50/M | 63 tokens/秒 | |
| 4 | Claude Fable 5.1 Anthropic | 83.15 | $10/M | $50/M | $10/M | $50/M | 62 tokens/秒 | |
| 5 | Claude Sonnet 5.5 Anthropic | 82.85 | $2/M | $10/M | $2/M | $10/M | 128 tokens/秒 | |
| 6 | Gemini 4 Argon 谷歌 | 81.23 | $2/M | $10/M | $2/M | $10/M | — | |
| 7 | Claude Fable 5 Anthropic | 81.22 | $10/M | $50/M | $10/M | $50/M | 65 tokens/秒 | |
| 8 | Claude Opus 5 Anthropic | 79.63 | $5/M | $25/M | $5/M | $25/M | 57 tokens/秒 | |
| 9 | GPT-6 Sol OpenAI | 78.32 | $2/M | $10/M | $2/M | $10/M | — | |
| 10 | GPT-5.6 Sol OpenAI | 75.25 | $4/M | $20/M | $4/M | $20/M | 97 tokens/秒 | |
| 11 | Muse Wonder Full Meta | 74.44 | — | — | — | — | — | |
| 12 | MiMo-V2.6-Pro 小米 | 73.53 | $0.435/M | $0.87/M | $0.435/M | $0.87/M | 47 tokens/秒 | |
| 13 | Claude Haiku 5.5 Anthropic | 71.54 | $0.1/M | $0.5/M | $0.1/M | $0.5/M | — | |
| 14 | Muse Wonder Meta | 71.07 | $4/M | $20/M | $4/M | $20/M | — | |
| 15 | GPT-5.6 Terra OpenAI | 70.26 | $2/M | $12/M | $2/M | $12/M | — | |
| 16 | Claude Opus 4.8 Anthropic | 70.12 | $5/M | $25/M | $5/M | $25/M | 64 tokens/秒 | |
| 17 | Kimi K3 月之暗面 | 69.78 | $3/M | $15/M | $0.192/M | $15/M | 52 tokens/秒 | |
| 18 | GPT-5.5 OpenAI | 69.42 | $5/M | $30/M | $5/M | $30/M | 104 tokens/秒 | |
| 19 | Step 5 Preview 阶跃星辰 | 69.42 | $1/M | $2.7/M | $1/M | $2.7/M | 87 tokens/秒 | |
| 20 | Qwen3.8 Max 阿里通义 | 69.30 | $2.0/M | $6.0/M | $1.475/M | $4.425/M | 37 tokens/秒 | |
| 21 | Gemini 3.8 Flash 谷歌 | 69.23 | $0.75/M | $3.75/M | $0.825/M | $4.125/M | 242 tokens/秒 | |
| 22 | Claude Opus 4.7 (Adaptive) Anthropic | 68.90 | $5/M | $25/M | — | — | — | |
| 23 | Ling 3.1 Flash 蚂蚁集团 InclusionAI | 68.04 | — | — | $0.075/M | $0.225/M | 211 tokens/秒 | |
| 24 | GLM-5.3 智谱 AI | 67.53 | 免费 | 免费 | $0.07/M | $1.4/M | 75 tokens/秒 | |
| 25 | Grok 4.6 xAI | 67.09 | $2/M | $6/M | $2/M | $6/M | 73 tokens/秒 | |
| 26 | Claude Opus 4.7 Anthropic | 66.96 | $5/M | $25/M | $5/M | $25/M | 50 tokens/秒 | |
| 27 | Gemini 3.7 Flash 谷歌 | 66.87 | $0.75/M | $3.75/M | $0.75/M | $3.75/M | 284 tokens/秒 | |
| 28 | GPT-5.4 OpenAI | 66.83 | $2.5/M | $15/M | $2.5/M | $15/M | 101 tokens/秒 | |
| 29 | GPT-6 Luna OpenAI | 66.64 | $0.1/M | $0.5/M | $0.1/M | $0.5/M | — | |
| 30 | Muse Spark 1.1 Meta | 65.92 | — | — | $1.25/M | $4.25/M | 217 tokens/秒 | |
| 31 | Muse Spark 1.2 Meta | 65.71 | $1.25/M | $4.25/M | $1.25/M | $4.25/M | 154 tokens/秒 | |
| 32 | Gemini 3.1 Pro 谷歌 | 64.99 | $2/M | $12/M | — | — | 125 tokens/秒 | |
| 33 | DeepSeek V4.1 Flash 深度求索 | 64.87 | $0.3/M | $1.2/M | $0.045/M | $0.5/M | 222 tokens/秒 | |
| 34 | Claude Sonnet 5 Anthropic | 64.59 | $2/M | $10/M | $2/M | $10/M | 86 tokens/秒 | |
| 35 | GPT-5.6 Luna OpenAI | 64.52 | $0.2/M | $1.2/M | $0.2/M | $1.2/M | — | |
| 36 | MiMo-V2.6-Flash 小米 | 64.34 | $0.14/M | $0.28/M | $0.14/M | $0.28/M | — | |
| 37 | Grok 4.5 xAI | 63.20 | $2/M | $6/M | $2/M | $6/M | 67 tokens/秒 | |
| 38 | DeepSeek V4 Pro 0813 深度求索 | 63.20 | $1.32/M | $3.96/M | $1.32/M | $3.96/M | 170 tokens/秒 | |
| 39 | Qwen3.8-Flash-Next 阿里通义 | 62.86 | 免费 | 免费 | — | — | 55 tokens/秒 | |
| 40 | Qwen3.7 Max 阿里通义 | 62.73 | $2.5/M | $7.5/M | $1.475/M | $4.425/M | 199 tokens/秒 | |
| 41 | Gemini 3.6 Flash 谷歌 | 62.52 | $0.75/M | $3.75/M | $0.75/M | $3.75/M | 197 tokens/秒 | |
| 42 | Gemini 3.5 Flash 谷歌 | 61.96 | $1.5/M | $9/M | $1.5/M | $9/M | 213 tokens/秒 | |
| 43 | Claude Opus 4.6 Anthropic | 61.48 | $5/M | $25/M | $5/M | $25/M | 42 tokens/秒 | |
| 44 | Hy4 preview 腾讯 | 61.45 | 免费 | 免费 | $0.834/M | $2.501/M | — | |
| 45 | Gemini 3 Pro 谷歌 | 60.88 | $2/M | $12/M | $1.25/M | $10/M | 109 tokens/秒 | |
| 46 | Kimi K2.6 月之暗面 | 60.52 | $0.95/M | $4/M | $0.65/M | $3.41/M | 59 tokens/秒 | |
| 47 | GPT-5.3 Codex OpenAI | 60.41 | $1.75/M | $14/M | $1.75/M | $14/M | 105 tokens/秒 | |
| 48 | GLM-5.2 智谱 AI | 60.37 | $1.4/M | $4.4/M | $0.006/M | $7/M | 93 tokens/秒 | |
| 49 | Muse Spark Meta | 60.22 | — | — | $1.25/M | $4.25/M | — | |
| 50 | GPT-5.2 OpenAI | 59.61 | $1.75/M | $14/M | $1.75/M | $14/M | 91 tokens/秒 | |
| 51 | Mistral Large 4 Mistral AI | 59.33 | $0.68/M | $2.09/M | $1.36/M | $4.18/M | 116 tokens/秒 | |
| 52 | Qwen3.8-27B 阿里通义 | 57.72 | 免费 | 免费 | $0.039/M | $1.25/M | 47 tokens/秒 | |
| 53 | Grok 4.20 xAI | 57.52 | $1.25/M | $2.5/M | $1.25/M | $2.5/M | 123 tokens/秒 | |
| 54 | GPT-5.1 OpenAI | 57.06 | $1.25/M | $10/M | $1.25/M | $10/M | 142 tokens/秒 | |
| 55 | GLM-5.3-Flash 智谱 AI | 56.87 | 免费 | 免费 | $0.046/M | $0.074/M | — | |
| 56 | Kimi K2.5 (Reasoning) 月之暗面 | 56.36 | $0.6/M | $3/M | — | — | — | |
| 57 | GLM-5.1 智谱 AI | 56.34 | $1.4/M | $4.4/M | $1.4/M | $4.4/M | 59 tokens/秒 | |
| 58 | Claude Sonnet 4.6 Anthropic | 55.83 | $3/M | $15/M | $3/M | $15/M | 49 tokens/秒 | |
| 59 | Qwen3.5 397B 阿里通义 | 55.54 | $0.6/M | $3.6/M | $0.17/M | $0.25/M | 90 tokens/秒 | |
| 60 | GPT-5.2-Codex OpenAI | 55.29 | $1.75/M | $14/M | $1.75/M | $14/M | 123 tokens/秒 | |
| 61 | MiniMax M3 MiniMax | 54.96 | $0.3/M | $1.2/M | $0.3/M | $1.2/M | 103 tokens/秒 | |
| 62 | GLM-5 智谱 AI | 54.84 | $1/M | $3.2/M | $1/M | $3.2/M | 78 tokens/秒 | |
| 63 | Gemini 3 Flash 谷歌 | 54.52 | $0.5/M | $3/M | $1.5/M | $9/M | 201 tokens/秒 | |
| 64 | Claude Opus 4.5 Thinking Anthropic | 54.18 | — | — | — | — | — | |
| 65 | Claude Opus 4.5 Anthropic | 54.13 | $5/M | $25/M | $5/M | $25/M | 50 tokens/秒 | |
| 66 | Qwen3.6 Plus 阿里通义 | 54.04 | $0.5/M | $3.0/M | $0.325/M | $1.95/M | 56 tokens/秒 | |
| 67 | Qwen3.7 Plus 阿里通义 | 53.86 | $0.4/M | $1.6/M | $0.32/M | $1.28/M | 56 tokens/秒 | |
| 68 | GPT-5.4 mini OpenAI | 53.47 | $0.75/M | $4.5/M | $0.75/M | $4.5/M | 201 tokens/秒 | |
| 69 | MAI-Thinking-1 微软 | 53.40 | — | — | — | — | — | |
| 70 | Grok 4.3 xAI | 53.21 | $1.25/M | $2.5/M | $1.25/M | $2.5/M | 139 tokens/秒 | |
| 71 | GLM-5-Turbo 智谱 AI | 53.19 | $1.2/M | $4/M | $1.2/M | $4/M | — | |
| 72 | Kimi K2.5 月之暗面 | 53.05 | $0.6/M | $3/M | $0.5/M | $2.5/M | 82 tokens/秒 | |
| 73 | MiMo-V2.5-Pro 小米 | 52.97 | — | — | $0.48/M | $1.8/M | — | |
| 74 | MiMo-V2-Pro 小米 | 52.88 | — | — | $0.48/M | $1.8/M | — | |
| 75 | Hy3 腾讯 | 52.51 | 免费 | 免费 | $0.132/M | $0.528/M | 88 tokens/秒 | |
| 76 | Hy3 Preview 腾讯 | 52.26 | 免费 | 免费 | $0.18/M | $0.6/M | — | |
| 77 | Grok 4 xAI | 51.89 | — | — | $1.25/M | $2.5/M | 54 tokens/秒 | |
| 78 | GPT-5.1-Codex OpenAI | 50.33 | $1.25/M | $10/M | $1.25/M | $10/M | — | |
| 79 | DeepSeek V3.2 深度求索 | 50.32 | $0.28/M | $0.42/M | $0.28/M | $0.42/M | 35 tokens/秒 | |
| 80 | MiMo-V2-Omni 小米 | 49.99 | — | — | — | — | — | |
| 81 | Kimi K2.7 Code 月之暗面 | 49.83 | $0.95/M | $4/M | $0.56/M | $3.1/M | 76 tokens/秒 | |
| 82 | Gemini 3.5 Flash-Lite 谷歌 | 49.83 | $0.3/M | $2.5/M | $0.3/M | $2.5/M | 347 tokens/秒 | |
| 83 | GLM-5V-Turbo 智谱 AI | 49.60 | $1.2/M | $4/M | $1.2/M | $4/M | — | |
| 84 | Gemini 2.5 Pro 谷歌 | 49.13 | $1.25/M | $10/M | $1.25/M | $10/M | 136 tokens/秒 | |
| 85 | Qwen3.5-27B 阿里通义 | 48.94 | 免费 | 免费 | $0.26/M | $2.6/M | 77 tokens/秒 | |
| 86 | Qwen3.6-27B 阿里通义 | 48.71 | 免费 | 免费 | $0.45/M | $2.7/M | 56 tokens/秒 | |
| 87 | GLM-4.7 智谱 AI | 48.69 | 免费 | 免费 | $0.54/M | $1.98/M | 45 tokens/秒 | |
| 88 | MiniMax M2.7 MiniMax | 48.15 | $0.3/M | $1.2/M | $0.3/M | $1.2/M | 76 tokens/秒 | |
| 89 | GPT-5.4 nano OpenAI | 47.37 | $0.2/M | $1.25/M | $0.2/M | $1.25/M | 191 tokens/秒 | |
| 90 | Nemotron 3 Ultra 英伟达 | 47.27 | 免费 | 免费 | — | — | — | |
| 91 | Gemini 3.1 Flash-Lite 谷歌 | 47.19 | $0.25/M | $1.5/M | $0.25/M | $1.5/M | 286 tokens/秒 | |
| 92 | Ling 3.0 Flash VL 蚂蚁集团 InclusionAI | 46.75 | 免费 | 免费 | $0.06/M | $0.18/M | — | |
| 93 | GPT-5 (medium) OpenAI | 46 | — | — | — | — | 83 tokens/秒 | |
| 94 | Qwen3.5-35B-A3B 阿里通义 | 46 | 免费 | 免费 | $0.14/M | $1/M | 147 tokens/秒 | |
| 95 | Grok 4.1 Fast (Reasoning) xAI | 45.88 | — | — | — | — | — | |
| 96 | Gemma 4 26B A4B 谷歌 | 45.80 | 免费 | 免费 | 免费 | 免费 | — | |
| 97 | o3 OpenAI | 45.63 | $2/M | $8/M | $2/M | $8/M | 137 tokens/秒 | |
| 98 | Ling 3.0 Flash 蚂蚁集团 InclusionAI | 45.28 | — | — | $0.021/M | $0.063/M | 344 tokens/秒 | |
| 99 | Step 3.7 Flash 阶跃星辰 | 43.82 | $0.2/M | $1.15/M | $0.2/M | $1.15/M | 172 tokens/秒 | |
| 100 | Qwen3.6-35B-A3B 阿里通义 | 43.81 | $0.248/M | $1.485/M | $0.1/M | $0.9/M | 135 tokens/秒 | |
| 101 | Qwen3.5 Flash 阿里通义 | 43.77 | $0.1/M | $0.4/M | $0.188/M | $1.125/M | — | |
| 102 | Grok 4 Fast (Reasoning) xAI | 43.52 | — | — | — | — | — | |
| 103 | Gemini 2.5 Flash 谷歌 | 42.23 | $0.3/M | $2.5/M | $0.3/M | $2.5/M | 186 tokens/秒 | |
| 104 | Qwen3.5-122B-A10B 阿里通义 | 41.84 | 免费 | 免费 | $0.26/M | $2.08/M | 129 tokens/秒 | |
| 105 | DeepSeek-R1 深度求索 | 41.38 | $0.55/M | $2.19/M | — | — | — | |
| 106 | Muse Glimmer 30B Meta | 41.35 | 免费 | 免费 | $0.3/M | $1.2/M | — | |
| 107 | Claude Haiku 4.5 Anthropic | 41.16 | $1/M | $5/M | $1/M | $5/M | — | |
| 108 | MiMo-V2-Flash 小米 | 41.08 | 免费 | 免费 | $0.14/M | $0.28/M | 129 tokens/秒 | |
| 109 | Grok 3 [Beta] xAI | 41.02 | — | — | — | — | — | |
| 110 | DeepSeek V3.1 深度求索 | 40.80 | 免费 | 免费 | $0.28/M | $0.42/M | — | |
| 111 | GLM-4.6 智谱 AI | 40.30 | — | — | $0.5/M | $2/M | 37 tokens/秒 | |
| 112 | o3-mini OpenAI | 40.21 | $1.1/M | $4.4/M | — | — | 160 tokens/秒 | |
| 113 | Gemma 4 31B 谷歌 | 40.01 | 免费 | 免费 | 免费 | 免费 | 36 tokens/秒 | |
| 114 | Qwen3 Max 阿里通义 | 39.55 | $1.2/M | $6.0/M | $1.475/M | $4.425/M | 64 tokens/秒 | |
| 115 | Claude 4.1 Opus Thinking Anthropic | 39.40 | — | — | — | — | 29 tokens/秒 | |
| 116 | DeepSeek V3.1 (Reasoning) 深度求索 | 39.07 | 免费 | 免费 | — | — | — | |
| 117 | o1 OpenAI | 38.58 | $15/M | $60/M | — | — | 98 tokens/秒 | |
| 118 | Claude 4 Sonnet Anthropic | 38.13 | $3/M | $15/M | $2/M | $10/M | 40 tokens/秒 | |
| 119 | LFM2.5-24B-A2B Liquid AI | 38.08 | $0.1/M | $1/M | $0.05/M | $0.7/M | — | |
| 120 | GPT-4.1 OpenAI | 37.57 | $2/M | $8/M | $2/M | $8/M | 165 tokens/秒 | |
| 121 | GPT-OSS 120B OpenAI | 37.05 | 免费 | 免费 | $0.15/M | $0.6/M | 197 tokens/秒 | |
| 122 | Gemma 4 12B 谷歌 | 36.95 | — | — | — | — | — | |
| 123 | DeepSeek V3 0324 深度求索 | 36.35 | — | — | $0.28/M | $0.42/M | — | |
| 124 | Command A+ Cohere | 36.29 | $2.5/M | $10/M | $2.5/M | $10/M | 139 tokens/秒 | |
| 125 | Grok 4.1 Fast xAI | 36.28 | $0.2/M | $0.5/M | — | — | 138 tokens/秒 | |
| 126 | Mercury 2.5 Inception | 36.22 | $0.04/M | $0.15/M | $0.04/M | $0.15/M | 691 tokens/秒 | |
| 127 | GLM-4.5-Air 智谱 AI | 34.90 | $0.2/M | $1.1/M | $0.13/M | $0.85/M | 69 tokens/秒 | |
| 128 | Mistral Small 4 Mistral AI | 34.67 | $0.15/M | $0.6/M | — | — | 178 tokens/秒 | |
| 129 | Grok Code Fast 1 xAI | 34.16 | $0.2/M | $1.5/M | — | — | 172 tokens/秒 | |
| 130 | Mistral Medium 3.5 128B Mistral AI | 33.84 | $1.5/M | $7.5/M | $0.4/M | $2/M | — | |
| 131 | Granite 4.2 8B IBM | 33.57 | 免费 | 免费 | $0.06/M | $0.15/M | 46 tokens/秒 | |
| 132 | Kimi K2 月之暗面 | 33.33 | $0.6/M | $2.5/M | $0.57/M | $2.3/M | 54 tokens/秒 | |
| 133 | GPT-OSS 20B OpenAI | 33.29 | 免费 | 免费 | $0.018/M | $0.09/M | 313 tokens/秒 | |
| 134 | Mistral Large 3 Mistral AI | 33.22 | $0.5/M | $1.5/M | $2/M | $6/M | 83 tokens/秒 | |
| 135 | Ling 2.6 Flash 蚂蚁集团 InclusionAI | 32.50 | — | — | — | — | 104 tokens/秒 | |
| 136 | Nemotron 3 Nano Omni 30B A3B 英伟达 | 32.42 | 免费 | 免费 | $0.06/M | $0.24/M | 268 tokens/秒 | |
| 137 | DeepSeek V3 深度求索 | 31.96 | $0.27/M | $1.1/M | $0.28/M | $0.42/M | — | |
| 138 | LFM2.5-2.6B Liquid AI | 31.28 | 免费 | 免费 | 免费 | 免费 | 197 tokens/秒 | |
| 139 | Gemma 4 E4B 谷歌 | 31.09 | 免费 | 免费 | — | — | — | |
| 140 | DeepSeek R1 Distill Qwen 32B 深度求索 | 30.86 | 免费 | 免费 | — | — | 60 tokens/秒 | |
| 141 | LFM2.5-8B-A1B Liquid AI | 30.20 | 免费 | 免费 | — | — | 335 tokens/秒 | |
| 142 | Gemma 4 E2B 谷歌 | 30.01 | 免费 | 免费 | — | — | — | |
| 143 | GPT-4o OpenAI | 29.67 | $2.5/M | $10/M | $2.5/M | $10/M | 151 tokens/秒 | |
| 144 | GPT-4.1 mini OpenAI | 29.46 | $0.4/M | $1.6/M | $0.4/M | $1.6/M | 80 tokens/秒 | |
| 145 | Llama 4 Scout Meta | 29.27 | 免费 | 免费 | $0.1/M | $0.3/M | 68 tokens/秒 | |
| 146 | Granite-4.0-H-1B IBM | 28.93 | 免费 | 免费 | — | — | — | |
| 147 | Mistral Medium 3 Mistral AI | 28.84 | $0.4/M | $2/M | $0.4/M | $2/M | 152 tokens/秒 | |
| 148 | Gemma 3 27B 谷歌 | 28.75 | 免费 | 免费 | $0.08/M | $0.35/M | 31 tokens/秒 | |
| 149 | Mistral Large 2 Mistral AI | 28.58 | — | — | $2/M | $6/M | 38 tokens/秒 | |
| 150 | Granite-4.0-H-350M IBM | 28.58 | 免费 | 免费 | — | — | — | |
| 151 | Granite-4.0-350M IBM | 28.58 | 免费 | 免费 | — | — | — | |
| 152 | Claude 3 Opus Anthropic | 28.55 | $15/M | $75/M | $5/M | $25/M | — | |
| 153 | Qwen3-Omni-30B-A3B-Instruct 阿里通义 | 28.16 | — | — | $0.15/M | $0.6/M | 96 tokens/秒 | |
| 154 | Gemini 1.5 Pro 谷歌 | 27.83 | $1.25/M | $5/M | $1.25/M | $10/M | — | |
| 155 | GPT-4o mini OpenAI | 26.86 | $0.15/M | $0.6/M | $0.15/M | $0.6/M | 33 tokens/秒 | |
| 156 | Nemotron 3.5 Lightning 30B A3B NVFP4 英伟达 | 26.82 | 免费 | 免费 | — | — | — | |
| 157 | Qwen2.5 Coder 32B Instruct 阿里通义 | 26.67 | 免费 | 免费 | $0.66/M | $1/M | — | |
| 158 | Phi-4 微软 | 25.13 | 免费 | 免费 | $0.07/M | $0.14/M | 40 tokens/秒 | |
| 159 | GPT-4.1 nano OpenAI | 24.56 | $0.1/M | $0.4/M | $0.1/M | $0.4/M | 181 tokens/秒 | |
| 160 | Llama 4 Maverick Meta | 22.08 | 免费 | 免费 | $0.188/M | $0.652/M | 66 tokens/秒 | |
| 161 | Claude 3 Haiku Anthropic | 14.18 | $0.25/M | $1.25/M | $1/M | $5/M | 138 tokens/秒 | |
| 162 | Gemini 1.0 Pro 谷歌 | 11.72 | — | — | — | — | — |