| 排名 | 模型 / 厂商 | 类型 | 综合 | 智能体 | 编程 | 推理 | 多模态 | 知识 | 多语言 | 指令 | 数学 | 输入 $/1M | 输出 $/1M | 上下文 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 Astra OpenAI |
闭源 | 88.69 | 70.58 | 74.61 | 89.5 | 82.9 | 86.98 | — | — | 85 | $10 | $50 | 1.1M |
| 2 | Claude Opus 5.5 Anthropic |
闭源 | 86.98 | 87.88 | 87.59 | 82.4 | 88.8 | 89.17 | — | — | — | $4 | $20 | 1M |
| 3 | Claude Fable 5.1 Anthropic |
闭源 | 83.00 | 78.68 | 81.26 | 81.4 | — | 86.06 | — | — | — | $10 | $50 | 1M |
| 4 | GPT-6 Sol OpenAI |
闭源 | 81.28 | 59.63 | 66 | 79.6 | 82.8 | 79.17 | — | — | — | $2 | $10 | 1.1M |
| 5 | Claude Opus 5 Anthropic |
闭源 | 79.81 | 77.35 | 72.8 | 77.3 | 88.8 | 81.4 | — | — | — | $5 | $25 | 1M |
| 6 | Claude Fable 5 Anthropic |
闭源 | 79.07 | 73.69 | 74.49 | 80.6 | 62.6 | 81.8 | — | 75.7 | — | $10 | $50 | 1M |
| 7 | GPT-5.6 Sol OpenAI |
闭源 | 78.49 | 69.56 | 71.55 | 72.1 | 87.6 | 78.82 | — | 87.7 | 96.8 | $4 | $20 | 1.1M |
| 8 | Gemini 3.8 Flash |
闭源 | 73.43 | 65.02 | 64.15 | 70.5 | 82.9 | 74.29 | — | — | — | $0.75 | $3.75 | 1M |
| 9 | GPT-5.6 Terra OpenAI |
闭源 | 72.58 | 59.47 | 64.73 | 65.4 | 77.2 | 71.13 | — | 85.7 | 96.8 | $2 | $12 | 1.1M |
| 10 | GPT-5.5 Pro OpenAI |
闭源 | 72.42 | — | — | 75 | — | — | — | — | 70.1 | $30 | $180 | 1M |
| 11 | Kimi K3 Moonshot AI |
待定 | 71.87 | 70.03 | 62.57 | 65.4 | 89.4 | 68.49 | — | — | — | $3 | $15 | — |
| 12 | Qwen3.8 Max Alibaba |
开源 | 71.80 | 65.55 | 56.01 | 87.6 | 87.4 | 66.69 | — | 90.5 | — | — | — | — |
| 13 | GPT-5.4 Pro OpenAI |
闭源 | 70.88 | — | — | 74.3 | — | — | — | — | 68.8 | $30 | $180 | 1.1M |
| 14 | Claude Opus 4.8 Anthropic |
闭源 | 70.46 | 61.3 | 63.32 | 59 | 87.7 | 70.04 | — | 74 | 65.2 | $5 | $25 | 1M |
| 15 | Grok 4.6 xAI |
闭源 | 69.19 | 67.9 | 62.02 | 57.6 | — | 69.02 | — | — | — | $2 | $6 | 500K |
| 16 | GPT-5.5 OpenAI |
闭源 | 69.13 | 60.07 | 64.11 | 65.9 | 71.4 | 69.97 | — | 91.9 | 69.4 | $5 | $30 | 1M |
| 17 | Claude Opus 4.7 (Adaptive) Anthropic |
闭源 | 68.56 | 59.13 | 57.8 | 53.6 | 50.1 | 64.14 | — | — | — | $5 | $25 | 1M |
| 18 | GPT-5.4 OpenAI |
闭源 | 68.51 | 49.35 | 50.71 | 60.5 | 69.3 | 66.82 | — | 89.2 | 64.4 | $2.5 | $15 | 1.1M |
| 19 | Gemini 3.7 Flash |
闭源 | 67.66 | 58.56 | 60.26 | 77.8 | 82.6 | 71.1 | — | — | — | $0.75 | $3.75 | 1M |
| 20 | Claude Sonnet 5 Anthropic |
闭源 | 67.01 | 64.16 | 60.81 | 78.4 | 77.4 | 64.49 | — | — | — | $2 | $10 | 1M |
| 21 | GPT-5.2 Pro OpenAI |
闭源 | 66.71 | — | — | — | — | — | — | — | — | $21 | $168 | 400K |
| 22 | GPT-6 Luna OpenAI |
闭源 | 66.59 | 55.55 | 53.69 | 55.2 | 71.3 | 66.56 | — | — | — | $0.1 | $0.5 | 1.1M |
| 23 | Claude Opus 4.7 Anthropic |
闭源 | 66.28 | 53.15 | 59.52 | — | — | 63.89 | — | — | 60.6 | $5 | $25 | 1M |
| 24 | Muse Spark 1.1 Meta |
闭源 | 65.92 | 57.71 | 56.29 | 75.5 | 77.3 | 68.16 | — | — | — | — | — | — |
| 25 | GPT-5.6 Luna OpenAI |
闭源 | 65.60 | 55.3 | 64.48 | 54.7 | 67.1 | 64.61 | — | — | 94.2 | $0.2 | $1.2 | 1.1M |
| 26 | GLM-5.3 Z.AI |
开源 | 65.55 | 67.17 | 57.77 | 76.8 | — | 61.69 | — | — | — | $0 | $0 | 1M |
| 27 | Grok 4.5 xAI |
闭源 | 65.27 | 57.88 | 58.1 | 50.3 | 78.5 | 66.81 | — | — | — | $2 | $6 | 500K |
| 28 | Muse Spark 1.2 Meta |
闭源 | 64.99 | 59.54 | 56.01 | 76.4 | — | 64.96 | — | — | — | $1.25 | $4.25 | — |
| 29 | Gemini 3.6 Flash |
闭源 | 64.58 | 45.29 | 54.62 | 63.9 | 82.6 | 64.33 | — | — | — | $0.75 | $3.75 | 1M |
| 30 | Gemini 3.1 Pro |
闭源 | 64.45 | 39.31 | 43.29 | 54.4 | 79.1 | 66.7 | — | — | 54.2 | $2 | $12 | 1M |
| 31 | Claude Opus 4.6 Anthropic |
闭源 | 63.64 | 44.29 | 51.05 | 68.1 | 59.6 | 58.68 | — | 51 | 58.5 | $5 | $25 | 1M |
| 32 | Gemini 3.5 Flash |
闭源 | 63.49 | 49.99 | 53.18 | 62.7 | 86.8 | 64.4 | — | 84.1 | 55.1 | $1.5 | $9 | 1M |
| 33 | DeepSeek V4 Pro 0813 DeepSeek |
开源 | 63.48 | 54.95 | 50.25 | 56.9 | — | 63.38 | — | — | 80.2 | $1.32 | $3.96 | 1M |
| 34 | Qwen3.7 Max Alibaba |
闭源 | 63.18 | 39.81 | 46.23 | 76.1 | — | 60.78 | 100 | 89.2 | 81.9 | — | — | — |
| 35 | dots3-note Preview Dots Studio |
开源 | 62.81 | 56.44 | — | 72.8 | 66.2 | — | — | 85.2 | — | $0 | $0 | 512K |
| 36 | GLM-5.2 Z.AI |
开源 | 62.52 | 55.63 | 57.17 | 75.9 | — | 57.2 | — | 88.5 | 80.7 | $1.4 | $4.4 | 1M |
| 37 | GPT-5.3 Codex OpenAI |
闭源 | 62.18 | 54.21 | 55.86 | 79.3 | 75.7 | 64.29 | — | 91.2 | — | $1.75 | $14 | 400K |
| 38 | Ornith-1.5-397B Ornith AI |
开源 | 61.76 | 56.33 | 54.9 | — | — | 63.29 | — | — | — | $0 | $0 | — |
| 39 | GPT-5.2 OpenAI |
闭源 | 61.37 | 42.57 | 40.95 | 60.6 | 66.3 | 57.9 | — | 91.2 | 57.4 | $1.75 | $14 | 400K |
| 40 | Gemini 3 Pro |
闭源 | 61.01 | — | 45.1 | 46.4 | 74.2 | 58.96 | — | 84.7 | 55.2 | $2 | $12 | 2M |
| 41 | Muse Spark Meta |
闭源 | 60.81 | 53 | 53.4 | 53.3 | 77.5 | 60.81 | — | 91.9 | 55.1 | — | — | — |
| 42 | Qwen3.8-Flash-Next Alibaba |
开源 | 60.72 | 57 | 55.57 | 76.8 | 83.1 | 57.28 | — | 87.2 | — | $0 | $0 | — |
| 43 | Grok 4.20 xAI |
闭源 | 59.66 | 24.53 | 23.78 | 41.1 | 34.6 | 50.96 | — | — | — | $1.25 | $2.5 | 1M |
| 44 | Hy4 preview Tencent |
开源 | 59.61 | 52.5 | 60.46 | — | 84.9 | 61.7 | — | — | — | $0 | $0 | — |
| 45 | Kimi K2.6 Moonshot AI |
开源 | 59.12 | 43.58 | 47.37 | — | 63.9 | 59.01 | — | — | 71 | $0.95 | $4 | — |
| 46 | GPT-5.1 OpenAI |
闭源 | 58.28 | — | 39.02 | 77.1 | 71.3 | 52.16 | — | 87.9 | 49 | $1.25 | $10 | 400K |
| 47 | GLM-5.3-Flash Z.AI |
开源 | 58.01 | 54.32 | 51.17 | — | 80.4 | 57 | — | — | — | $0 | $0 | 1M |
| 48 | GLM-5.1 Z.AI |
开源 | 57.14 | 41.93 | 50.88 | 72.7 | — | 52.28 | — | 92.4 | 63.8 | $1.4 | $4.4 | 203K |
| 49 | Claude Sonnet 4.6 Anthropic |
闭源 | 56.35 | 41.79 | 48.14 | 69 | 54.1 | 53.03 | — | 46.6 | 48.8 | $3 | $15 | 200K |
| 50 | Qwen3.7 Plus Alibaba |
闭源 | 55.78 | 35.16 | 43.56 | 75.1 | 72.4 | 52.3 | 78.9 | 89.2 | 78.2 | — | — | — |
| 51 | Inkling-Small Thinking Machines Lab |
开源 | 55.63 | 35.6 | 40.25 | 51.3 | 48.8 | 54.05 | — | 89.2 | 76.5 | $0.58 | $1.44 | — |
| 52 | Gemini 3 Flash |
闭源 | 55.61 | 31.81 | 36.45 | 60.1 | 75.8 | 54.51 | — | 64.7 | 50 | $0.5 | $3 | 1M |
| 53 | Qwen3.8-27B Alibaba |
开源 | 55.35 | 61.03 | 49.42 | 78.4 | 79.8 | 50.2 | — | 83.2 | — | $0 | $0 | — |
| 54 | Claude Opus 4.5 Anthropic |
闭源 | 55.09 | 30.88 | 43.49 | 73.2 | 23.5 | 49.44 | 82.9 | 30.6 | 57.9 | $5 | $25 | 200K |
| 55 | Claude Opus 4.5 Thinking Anthropic |
闭源 | 55.07 | — | — | 75.2 | 69.1 | 53.93 | — | 68.5 | — | — | — | — |
| 56 | GPT-5.4 mini OpenAI |
闭源 | 55.01 | 33.91 | 37.24 | 39.9 | 57.3 | 48.37 | — | 88.5 | 44.3 | $0.75 | $4.5 | 400K |
| 57 | Qwen 3.6 Max (preview) Alibaba |
闭源 | 54.97 | — | 46.93 | — | — | — | — | — | 41.3 | — | — | — |
| 58 | MiniMax M3 MiniMax |
开源 | 54.86 | 39.87 | 39.24 | 79.1 | 52 | 49.34 | — | 92.4 | — | $0.3 | $1.2 | — |
| 59 | Grok 4.3 xAI |
闭源 | 54.73 | 25.58 | 29.54 | 66.3 | 65.7 | 52.96 | — | 93.2 | — | $1.25 | $2.5 | 1M |
| 60 | Qwen3.6 Plus Alibaba |
闭源 | 54.59 | 33.24 | 43.31 | 61.2 | 66.2 | 52.05 | 69.7 | 82.5 | 62 | — | — | — |
| 61 | GLM-5 Z.AI |
开源 | 54.45 | 37.17 | 39.27 | 53.2 | — | 47.07 | 48.7 | 87.2 | 56.5 | $1 | $3.2 | 200K |
| 62 | Inkling Thinking Machines Lab |
开源 | 54.17 | 36.19 | 36.74 | 75.2 | 47.6 | 53.51 | — | 83.9 | 78.5 | $1.87 | $4.68 | — |
| 63 | Kimi K2.5 Moonshot AI |
开源 | 52.96 | 36.07 | 38.87 | 55.1 | 65.8 | 47.79 | 38.2 | 84.4 | 62.1 | $0.6 | $3 | — |
| 64 | GLM-5-Turbo Z.AI |
闭源 | 52.76 | — | — | 71.4 | — | 45.54 | — | 88.3 | — | $1.2 | $4 | 200K |
| 65 | Grok 4 xAI |
闭源 | 52.72 | — | — | 68.8 | 61.4 | 48.59 | — | 62.9 | 38.3 | — | — | 128K |
| 66 | MiMo-V2-Pro Xiaomi |
闭源 | 52.60 | — | 38.58 | 69 | — | 46.1 | — | 82.6 | — | — | — | — |
| 67 | Ternary Bonsai 2 27B Prism ML |
开源 | 52.04 | — | — | 75 | — | — | — | 71 | 76.8 | $0 | $0 | — |
| 68 | MiMo-V2.5-Pro Xiaomi |
闭源 | 52.03 | 37.47 | 45.7 | 76.8 | — | 49.84 | — | 92.4 | — | — | — | — |
| 69 | Hy3 Tencent |
开源 | 51.45 | 35.21 | 51.16 | 76.4 | — | 48.12 | — | — | — | $0 | $0 | — |
| 70 | Gemini 3.5 Flash-Lite |
闭源 | 50.96 | 30.4 | 39.49 | 61.1 | 76.4 | 49.1 | — | — | — | $0.3 | $2.5 | 1M |
| 71 | GPT-5.4 nano OpenAI |
闭源 | 50.81 | 31.69 | 31.56 | 36.9 | 23.8 | 37.75 | — | 91.9 | 43.8 | $0.2 | $1.25 | 400K |
| 72 | Kimi K2.7 Code Moonshot AI |
开源 | 50.44 | 37.1 | 45.05 | 76.6 | — | 53.22 | — | 75.1 | — | $0.95 | $4 | — |
| 73 | Gemini 2.5 Pro |
闭源 | 50.24 | 24.73 | 24.56 | 69.5 | 70.4 | 44.23 | — | 56.4 | 35.1 | $1.25 | $10 | 1M |
| 74 | DeepSeek V3.2 DeepSeek |
开源 | 50.21 | — | 32.27 | 53.5 | — | 41.59 | — | 56.7 | 40 | $0.28 | $0.42 | 128K |
| 75 | GPT-5.2-Codex OpenAI |
闭源 | 50.15 | 34.23 | 47.14 | 78.6 | 72.5 | 51.98 | — | 92.4 | — | $1.75 | $14 | 400K |
| 76 | MiniMax M2.5 MiniMax |
闭源 | 49.89 | 25.38 | 35.89 | — | — | — | — | — | — | $0.3 | $1.2 | — |
| 77 | GLM-4.7 Z.AI |
开源 | 49.23 | 27.02 | 39.62 | 70.9 | — | 40.56 | — | 81.4 | 25.8 | $0 | $0 | 200K |
| 78 | MiMo-V2-Omni Xiaomi |
闭源 | 49.22 | — | — | 73.6 | 63.1 | 41.69 | — | 62.6 | — | — | — | — |
| 79 | Apodex 1.1 Apodex |
闭源 | 48.97 | 44.25 | 43.17 | 76.6 | 76.7 | 53.24 | — | — | — | — | — | — |
| 80 | GLM-5V-Turbo Z.AI |
闭源 | 48.89 | — | 34.41 | 70.4 | 67.3 | 43.53 | — | 72.5 | — | $1.2 | $4 | 200K |
| 81 | Qwen3.5 Plus Alibaba |
闭源 | 48.71 | — | — | — | — | — | — | — | 39.3 | $0.4 | $2.4 | — |
| 82 | Gemini 3.1 Flash-Lite |
闭源 | 48.51 | 28.75 | 27.3 | — | 37.4 | 47.38 | — | — | — | $0.25 | $1.5 | 1M |
| 83 | MiniMax M2.7 MiniMax |
开源 | 48.08 | 25.2 | 36.14 | 75.9 | — | 42.98 | — | 91.6 | — | $0.3 | $1.2 | — |
| 84 | Claude Sonnet 4.5 Anthropic |
闭源 | 47.85 | 34.13 | — | 19.1 | — | — | — | — | 34.6 | $3 | $15 | 200K |
| 85 | Qwen3.7 Flash Alibaba |
闭源 | 47.37 | — | — | — | — | — | — | — | — | $0.03 | $0.13 | — |
| 86 | o3-pro OpenAI |
闭源 | 47.25 | — | — | — | — | — | — | — | — | $20 | $80 | — |
| 87 | Qwen3.5-35B-A3B Alibaba |
开源 | 47.05 | 28.94 | 26.76 | 41.3 | 67.2 | 39.54 | 21.1 | 87.4 | — | $0 | $0 | — |
| 88 | o3 OpenAI |
闭源 | 47.04 | — | — | 73.4 | 63.3 | 47.15 | — | 86 | 37.7 | $2 | $8 | — |
| 89 | Qwen3.6-27B Alibaba |
开源 | 46.49 | 28.08 | 37.7 | 75.2 | 51.9 | 45.66 | — | 81 | 72.1 | $0 | $0 | — |
| 90 | Gemma 4 26B A4B |
开源 | 46.28 | — | 34.33 | 67.2 | 44.2 | 36.43 | — | 87.3 | — | $0 | $0 | 256K |
| 91 | Grok 4.1 Fast (Reasoning) xAI |
闭源 | 45.68 | — | 27.57 | 72.9 | 53.3 | 41.35 | — | 61.6 | — | — | — | — |
| 92 | Hy3 Preview Tencent |
开源 | 45.64 | — | 30.7 | — | — | 40.41 | — | 46.9 | — | $0 | $0 | — |
| 93 | Qwen3.5 Flash Alibaba |
闭源 | 45.49 | — | 26.47 | — | — | 42.98 | — | — | 28.4 | $0.1 | $0.4 | — |
| 94 | Ling 3.0 Flash InclusionAI |
开源 | 45.35 | 32.95 | 34.93 | 72.2 | — | 43.08 | — | 72.1 | 72.9 | — | — | — |
| 95 | Grok 3 [Beta] xAI |
闭源 | 45.04 | — | — | — | — | — | — | — | 26.8 | — | — | 128K |
| 96 | GPT-5 (medium) OpenAI |
闭源 | 44.92 | — | 34.64 | 74.3 | 69.5 | 46.6 | — | 84.9 | — | — | — | 128K |
| 97 | Gemma 4 31B |
开源 | 44.84 | 23.88 | 35.74 | 70 | 58.5 | 40.04 | — | 91.5 | — | $0 | $0 | 256K |
| 98 | GPT-5.1-Codex OpenAI |
闭源 | 44.47 | — | 33.06 | 69.7 | 66.9 | 48.24 | — | 84.2 | — | $1.25 | $10 | 400K |
| 99 | GPT-5 mini OpenAI |
闭源 | 44.05 | 27.41 | 30.04 | — | — | — | — | — | — | $0.25 | $2 | 128K |
| 100 | Qwen3.5-27B Alibaba |
开源 | 43.81 | 32.2 | 32.73 | 52.6 | 70.6 | 42.27 | 36.8 | 91.5 | — | $0 | $0 | — |
| 101 | Grok 3 Mini xAI |
闭源 | 43.64 | — | — | — | — | — | — | — | — | $0.3 | $0.5 | 128K |
| 102 | Nemotron 3 Ultra NVIDIA |
开源 | 43.30 | 25.49 | 23.17 | 49.8 | — | 45.9 | 47.4 | 88.1 | — | $0 | $0 | — |
| 103 | Gemini 2.5 Flash |
闭源 | 43.00 | — | — | 56.4 | 56.6 | 35.73 | — | 43.7 | 28.6 | $0.3 | $2.5 | 1M |
| 104 | DeepSeek-R1 DeepSeek |
开源 | 42.63 | — | — | 60.4 | — | 38.72 | — | 44.5 | — | $0.55 | $2.19 | — |
| 105 | Claude Haiku 4.5 Anthropic |
闭源 | 42.48 | 21.74 | 19.78 | — | — | 34.5 | — | — | 28.8 | $1 | $5 | 200K |
| 106 | Grok 4 Fast (Reasoning) xAI |
闭源 | 42.43 | — | 23.76 | 72.7 | 51.1 | 39.67 | — | 58.7 | — | — | — | — |
| 107 | Qwen3.6-35B-A3B Alibaba |
开源 | 41.88 | 29.6 | 36.39 | 71.4 | 50 | 42.13 | — | 76.8 | 70.7 | — | — | — |
| 108 | Muse Glimmer 30B Meta |
开源 | 41.73 | 27.61 | 36.28 | 79.3 | 46.3 | 43.85 | — | 77.7 | 75.4 | $0 | $0 | — |
| 109 | Step 3.7 Flash StepFun |
开源 | 41.41 | 35.16 | 32.38 | 72.7 | 71 | 43.05 | — | 80.6 | — | $0.2 | $1.15 | — |
| 110 | DeepSeek V3.1 DeepSeek |
开源 | 41.27 | — | — | 54.4 | — | 36.07 | — | 42.1 | — | $0 | $0 | 128K |
| 111 | Step 3.5 Flash StepFun |
开源 | 41.12 | — | — | — | — | — | — | — | — | $0.1 | $0.3 | — |
| 112 | o3-mini OpenAI |
闭源 | 40.16 | — | — | — | — | 35.33 | — | — | — | $1.1 | $4.4 | — |
| 113 | Qwen3.5-122B-A10B Alibaba |
开源 | 40.06 | 23.09 | 35.73 | 49.8 | 57 | 41.5 | 36.8 | 91.6 | — | $0 | $0 | — |
| 114 | MiMo-V2-Flash Xiaomi |
开源 | 39.92 | — | 32.45 | 47 | — | 33.04 | — | 44.9 | — | $0 | $0 | — |
| 115 | GLM-4.6 Z.AI |
开源 | 39.80 | — | 27.66 | 40.2 | — | 35.92 | — | 40.7 | 27.3 | — | — | — |
| 116 | GPT-4.1 OpenAI |
闭源 | 39.79 | — | — | 69 | 50.3 | 35.48 | — | 48.9 | 28 | $2 | $8 | 1M |
| 117 | Qwen3 Max Alibaba |
闭源 | 39.11 | — | — | 56.4 | — | 38.89 | — | 50.3 | — | — | — | — |
| 118 | Claude 4.1 Opus Anthropic |
闭源 | 38.47 | — | — | — | — | — | — | — | — | $15 | $75 | 200K |
| 119 | DeepSeek V3.1 (Reasoning) DeepSeek |
开源 | 38.43 | — | — | 61 | — | 38.47 | — | 47 | — | $0 | $0 | 128K |
| 120 | GLM-4.7-Flash Z.AI |
开源 | 38.16 | — | — | — | — | — | — | — | — | $0 | $0 | 200K |
| 121 | GPT-OSS 120B OpenAI |
开源 | 37.76 | 19.66 | 26.67 | 57.8 | — | 36.86 | — | 82.8 | — | $0 | $0 | 128K |
| 122 | o1 OpenAI |
闭源 | 37.58 | — | 30.6 | 66.7 | — | 40.78 | — | 84.5 | 32.5 | $15 | $60 | — |
| 123 | K-Exaone LG AI Research |
闭源 | 37.08 | — | 27.87 | 64.2 | — | 35.72 | — | 77.2 | — | — | — | — |
| 124 | Claude 4.1 Opus Thinking Anthropic |
闭源 | 36.62 | — | — | 74.3 | 60.1 | 41.52 | — | 65.1 | — | — | — | 200K |
| 125 | Mercury 2 Inception |
闭源 | 36.43 | 17.74 | 19.07 | — | — | — | — | — | — | $0.25 | $0.75 | — |
| 126 | Grok 4.1 Fast xAI |
闭源 | 36.33 | — | — | 43.6 | 31.4 | 32.19 | — | 40.4 | — | $0.2 | $0.5 | 2M |
| 127 | o1-preview OpenAI |
闭源 | 36.32 | — | 23.9 | — | — | — | — | — | — | $15 | $60 | — |
| 128 | Command A+ Cohere |
开源 | 36.17 | 19.21 | 26.32 | 58.3 | 15.1 | 29.53 | — | 89.2 | — | $2.5 | $10 | 128K |
| 129 | Claude 4 Sonnet Anthropic |
闭源 | 36.08 | — | — | 52.3 | 52 | 36.93 | — | 52 | — | $3 | $15 | 200K |
| 130 | Mistral Medium 3.5 128B Mistral |
开源 | 36.05 | 20.05 | 26.86 | 69.7 | 55.7 | 32.14 | — | 82.6 | — | $1.5 | $7.5 | — |
| 131 | o1-pro OpenAI |
闭源 | 35.65 | — | — | — | — | — | — | — | — | $150 | $600 | — |
| 132 | GPT-5 nano OpenAI |
闭源 | 35.01 | — | — | — | — | — | — | — | — | $0.05 | $0.4 | 400K |
| 133 | GLM-4.5-Air Z.AI |
闭源 | 34.64 | — | — | 54.2 | — | 33 | — | 41.9 | — | $0.2 | $1.1 | 128K |
| 134 | Mistral Small 4 Mistral |
开源 | 34.29 | 17.42 | 24.99 | 56.2 | 43.8 | 34.68 | — | 55.7 | — | $0.15 | $0.6 | — |
| 135 | Trinity-Large-Thinking Arcee AI |
开源 | 34.25 | 17.05 | 21.15 | 48.2 | — | 35.63 | — | 66.3 | — | $0.25 | $0.9 | — |
| 136 | Mistral Large 3 Mistral |
闭源 | 34.13 | 16.31 | 18.57 | 46.8 | 42.2 | 33.44 | — | 40 | — | $0.5 | $1.5 | — |
| 137 | Ling 2.6 Flash InclusionAI |
开源 | 33.65 | — | 23.79 | 43.6 | — | 31.49 | — | 37.5 | — | — | — | — |
| 138 | GPT-OSS 20B OpenAI |
开源 | 33.59 | 15.24 | 21.97 | 45.9 | — | 32.68 | — | 77.8 | — | $0 | $0 | 128K |
| 139 | Gemma 4 E4B |
开源 | 33.10 | — | 20.37 | 44.1 | 35.8 | 29.36 | — | 50.5 | — | $0 | $0 | 128K |
| 140 | Sarvam 105B Sarvam |
开源 | 33.01 | — | — | 22.1 | — | 32.35 | — | 37.7 | — | $0 | $0 | — |
| 141 | Ornith-1.5-35B-A3B Ornith AI |
开源 | 32.72 | 24.02 | 24.7 | — | — | 34.15 | — | — | — | $0 | $0 | — |
| 142 | LFM2.5-2.6B LiquidAI |
开源 | 32.72 | 19.69 | 19.75 | 26 | — | 27.86 | — | 38.1 | — | $0 | $0 | — |
| 143 | Gemma 4 E2B |
开源 | 32.27 | — | 19.3 | 33.3 | 25.8 | 27.55 | — | 42.4 | — | $0 | $0 | 128K |
| 144 | Grok Code Fast 1 xAI |
闭源 | 32.05 | — | 21.76 | 58.5 | — | 35.85 | — | 46.8 | — | $0.2 | $1.5 | 256K |
| 145 | DeepSeek V3 DeepSeek |
开源 | 31.90 | 11.12 | 18.94 | 42.2 | — | 29.97 | — | 38.2 | 26 | $0.27 | $1.1 | 128K |
| 146 | LFM2.5-8B-A1B LiquidAI |
开源 | 31.87 | — | — | 22.1 | — | 27.16 | — | 35.4 | 17.3 | $0 | $0 | — |
| 147 | Solar Pro 2 Upstage |
闭源 | 31.71 | — | — | 22.1 | — | 30.14 | — | 36.8 | — | — | — | — |
| 148 | Sarvam 30B Sarvam |
开源 | 31.39 | — | — | 22.1 | — | 29.52 | — | 27.4 | — | $0 | $0 | — |
| 149 | Qwen2.5-72B Alibaba |
开源 | 31.34 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 150 | Exaone 4.0 32B LG AI Research |
开源 | 31.22 | — | — | — | — | 29.04 | — | 36.5 | — | — | — | — |
| 151 | ZAYA1-8B Zyphra |
开源 | 31.10 | — | — | — | — | 32.34 | — | 23.5 | 61.3 | $0 | $0 | — |
| 152 | Qwen3-Omni-30B-A3B-Instruct Alibaba |
开源 | 30.99 | — | — | 22.1 | 41.9 | 29.4 | — | 33.5 | — | — | — | — |
| 153 | Nemotron 3 Nano Omni 30B A3B NVIDIA |
开源 | 30.84 | — | 20.7 | 49.4 | 39.3 | 31.88 | — | 72.4 | — | $0 | $0 | — |
| 154 | Llama 3.1 405B Meta |
开源 | 30.77 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 155 | GPT-4o OpenAI |
闭源 | 30.77 | — | — | 56 | — | 31.46 | — | 37.6 | 25.1 | $2.5 | $10 | 128K |
| 156 | DeepSeek R1 Distill Qwen 32B DeepSeek |
开源 | 30.45 | — | — | 28.1 | — | 28.39 | — | 27.4 | — | $0 | $0 | 128K |
| 157 | Exaone 4.0 1.2B LG AI Research |
开源 | 30.43 | — | — | 22.1 | — | 26.27 | — | 27.4 | — | — | — | — |
| 158 | Granite-4.0-H-1B IBM |
开源 | 30.41 | — | — | 26.9 | — | 25.86 | — | 27.4 | — | $0 | $0 | 128K |
| 159 | Granite 4.2 8B IBM |
开源 | 30.20 | 13.48 | 18.61 | 53 | — | 28 | — | 82.8 | — | $0 | $0 | 128K |
| 160 | Granite-4.0-350M IBM |
开源 | 30.15 | — | — | 22.1 | — | 25.12 | — | 27.4 | — | $0 | $0 | 32K |
| 161 | Granite-4.0-H-350M IBM |
开源 | 30.15 | — | — | 22.1 | — | 25.2 | — | 27.4 | — | $0 | $0 | 32K |
| 162 | Claude 3.5 Sonnet Anthropic |
闭源 | 29.95 | — | — | — | — | — | — | — | 25.6 | $3 | $15 | 200K |
| 163 | Llama 4 Scout Meta |
开源 | 29.42 | 8.9 | 14.26 | 41.1 | 38.1 | 26.16 | — | 44.3 | 25.1 | $0 | $0 | — |
| 164 | GPT-4.1 mini OpenAI |
闭源 | 29.13 | — | 20.03 | 52.3 | 46.6 | 30.96 | — | 42.8 | 28.4 | $0.4 | $1.6 | 1M |
| 165 | Mistral Large 2 Mistral |
闭源 | 29.02 | — | — | — | — | 31.29 | — | 33.5 | — | — | — | — |
| 166 | Gemma 3 27B |
开源 | 28.74 | 6.93 | 12.97 | 27.1 | 30.8 | 24.07 | — | 34.3 | — | $0 | $0 | 32K |
| 167 | Claude 3 Opus Anthropic |
闭源 | 28.66 | — | 14.86 | — | — | 23.36 | — | — | — | $15 | $75 | 200K |
| 168 | Gemma 4 12B |
开源 | 28.47 | — | 22.04 | 34.7 | 26.2 | 31.07 | — | 88.7 | 53 | — | — | — |
| 169 | Mistral Medium 3 Mistral |
闭源 | 28.30 | — | — | 43.6 | 38.2 | 29.68 | — | 44.1 | — | $0.4 | $2 | — |
| 170 | Ornith-1.5-9B Ornith AI |
开源 | 28.16 | 17.99 | — | — | — | 29.49 | — | — | — | $0 | $0 | — |
| 171 | Gemini 1.5 Pro |
闭源 | 27.93 | — | 22.44 | — | 41.1 | 30.05 | — | — | — | $1.25 | $5 | 1M |
| 172 | Qwen2.5 Coder 32B Instruct Alibaba |
开源 | 27.12 | — | — | — | — | 28.36 | — | — | — | $0 | $0 | — |
| 173 | GPT-4o mini OpenAI |
闭源 | 26.88 | — | 12.03 | — | 25.4 | 20.55 | — | 33.2 | — | $0.15 | $0.6 | 128K |
| 174 | Kimi K2 Moonshot AI |
闭源 | 26.42 | — | — | 58.5 | — | 37.15 | — | 47 | 39 | $0.6 | $2.5 | — |
| 175 | Phi-4 Microsoft |
开源 | 25.93 | — | — | 22.1 | — | 25.65 | — | 27.4 | — | $0 | $0 | — |
| 176 | Ministral 3 14B Mistral |
开源 | 25.27 | 13.84 | 19.61 | — | — | — | — | — | — | $0.2 | $0.2 | — |
| 177 | MiniMax M1 80k MiniMax |
闭源 | 24.88 | — | — | — | — | — | — | — | — | — | — | — |
| 178 | GPT-4.1 nano OpenAI |
闭源 | 24.87 | — | 15.11 | 36 | 25.4 | 24.96 | — | 34.6 | 25.4 | $0.1 | $0.4 | 1M |
| 179 | Llama 3 70B Meta |
开源 | 23.58 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 180 | Llama 4 Maverick Meta |
开源 | 22.96 | 11.88 | 18.36 | 56.4 | 51.6 | 30.73 | — | 48.9 | 25.1 | $0 | $0 | — |
| 181 | GPT-4 Turbo OpenAI |
闭源 | 21.64 | — | 14.06 | — | — | — | — | — | — | $10 | $30 | 128K |
| 182 | Mixtral 8x22B Instruct v0.1 Mistral |
开源 | 21.12 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 183 | Nova Pro Amazon |
闭源 | 20.79 | — | — | 36.5 | 25.4 | 29.63 | — | 42.5 | — | — | — | — |
| 184 | LFM2-24B-A2B LiquidAI |
闭源 | 19.89 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 185 | Ministral 3 8B Mistral |
开源 | 19.28 | 13.34 | 18.41 | — | — | — | — | — | — | $0.15 | $0.15 | — |
| 186 | Nemotron 3.5 Lightning 30B A3B NVFP4 NVIDIA |
开源 | 19.10 | 7.27 | 11.4 | 55.9 | — | 20.65 | — | 68.5 | — | $0 | $0 | — |
| 187 | LFM2.5-1.2B-Thinking LiquidAI |
闭源 | 18.61 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 188 | LFM2.5-1.2B-Instruct LiquidAI |
闭源 | 18.30 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 189 | Ministral 3 3B Mistral |
开源 | 18.05 | 12.96 | 17.14 | — | — | — | — | — | — | $0.1 | $0.1 | — |
| 190 | Claude 3 Haiku Anthropic |
闭源 | 14.75 | — | — | 41.1 | 25.4 | 19.7 | — | 39.9 | — | $0.25 | $1.25 | 200K |
| 191 | Gemini 1.0 Pro |
闭源 | 12.95 | — | — | — | — | 15.41 | — | — | — | — | — | 32K |
| 192 | Nemotron-4 15B NVIDIA |
开源 | 11.42 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 193 | Mistral 7B v0.3 Mistral |
开源 | 8.14 | — | — | — | — | — | — | — | — | $0 | $0 | — |
| 194 | MiniCPM5-1B OpenBMB |
开源 | 4.10 | — | — | — | — | 4.58 | — | 10.4 | 5.4 | — | — | — |
当前:综合榜 · 共 194 个模型 · 按维度排序时缺测评数据的模型排在末尾
分数为第三方测评汇总,仅供参考 · 数据更新于 2026-09-27 01:13 · 来自 BenchLM.ai · 价格单位:美元/百万 tokens
分数为第三方测评汇总,仅供参考 · 数据更新于 2026-09-27 01:13 · 来自 BenchLM.ai · 价格单位:美元/百万 tokens
性价比排行 每一块钱能买到多少分 · 价格缺失的模型不参与
性价比 = 综合分 ÷ 平均价((输入价 + 输出价) ÷ 2),单位:分 / 美元
价格 vs 综合分 对数价格轴 · 越靠左上越"又便宜又强"
能力维度曲线对比 默认综合榜前 4 名 · 缺测评数据的维度会出现断点
可下拉更换对比模型
模型能力对比 六个核心维度 · 下拉切换模型即时重绘 · 缺数据维度按 0 绘制
VS
数据更新于 2026-09-27 01:13 · 来自 BenchLM.ai(本站每日自动同步)· 分数为第三方测评汇总仅供参考 · 性价比与图表由浏览器端根据本站最新数据实时计算
