LOADING加载中
154 个模型
| 排名 | 模型 | 评分 | 评分下界 | 评分上界 | 方差 | 投票数 | 分类 | 源数据更新 |
|---|---|---|---|---|---|---|---|---|
| #1 | Claude Opus 5.5 (Max)anthropic · Proprietary | 1,805.65 | 1,829.75 | 37.79 | 4,009 | overall | 2026-11-09 | |
| #2 | Claude Sonnet 5.5 (xHigh)anthropic · Proprietary | 1,763.96 | 1,784.55 | 27.58 | 5,493 | overall | 2026-11-09 | |
| #3 | Gemini 4 Carbon (High)google · Proprietary | 1,752.68 | 1,783.28 | 60.94 | 2,128 | overall | 2026-11-09 | |
| #4 | GPT-6 Astra (Max)openai · Proprietary | 1,758.02 | 1,775.65 | 20.22 | 8,117 | overall | 2026-11-09 | |
| #5 | Claude Fable 5.1 (Max)anthropic · Proprietary | 1,736.45 | 1,752.67 | 17.13 | 9,453 | overall | 2026-11-09 | |
| #6 | GPT-6.1 Sol (Max)openai · Proprietary | 1,732.72 | 1,752.96 | 26.67 | 5,250 | overall | 2026-11-09 | |
| #7 | Muse Wonder Full (Max)meta · Proprietary | 1,709.52 | 1,738.27 | 53.81 | 2,080 | overall | 2026-11-09 | |
| #8 | Claude Sonnet 5.5 (High)anthropic · Proprietary | 1,704.63 | 1,721.84 | 19.27 | 7,541 | overall | 2026-11-09 | |
| #9 | Claude Opus 5 (Max)anthropic · Proprietary | 1,684.74 | 1,696.94 | 9.68 | 19,593 | overall | 2026-11-09 | |
| #10 | Kimi K3.1 Pro (Max)moonshot · Kimi K3.1 Pro License | 1,677.82 | 1,697.88 | 26.19 | 4,591 | overall | 2026-11-09 | |
| #11 | Gemini 4 Argon (High)google · Proprietary | 1,672.55 | 1,692.63 | 26.24 | 4,626 | overall | 2026-11-09 | |
| #12 | GPT-6.1 Terra (Max)openai · Proprietary | 1,667.52 | 1,689.95 | 32.76 | 3,559 | overall | 2026-11-09 | |
| #13 | Qwen3.8-Max (0902)alibaba · Proprietary | 1,670.36 | 1,683.04 | 10.46 | 21,354 | overall | 2026-11-09 | |
| #14 | GPT-6 Sol (Max)openai · Proprietary | 1,665.26 | 1,682.44 | 19.22 | 6,882 | overall | 2026-11-09 | |
| #15 | Qwen3.8-Maxalibaba · Proprietary | 1,658.2 | 1,682.41 | 38.14 | 3,373 | overall | 2026-11-09 | |
| #16 | DeepSeek-V4.1-Flash-20261017 (Max)deepseek · MIT | 1,657.35 | 1,671.05 | 12.2 | 13,241 | overall | 2026-11-09 | |
| #17 | Kimi K3 (Max)moonshot · Kimi K3 license | 1,651.32 | 1,662.79 | 8.57 | 22,114 | overall | 2026-11-09 | |
| #18 | Claude Opus 5 (High)anthropic · Proprietary | 1,651.33 | 1,661.88 | 7.25 | 26,982 | overall | 2026-11-09 | |
| #19 | Muse Spark 1.3 (Max)meta · Proprietary | 1,647.15 | 1,661.4 | 13.21 | 11,758 | overall | 2026-11-09 | |
| #20 | Muse Wonder (Max)meta · Proprietary | 1,640.52 | 1,660.03 | 24.76 | 4,875 | overall | 2026-11-09 | |
| #21 | Inkling 2 Smallthinky · Proprietary | 1,629.78 | 1,669.16 | 100.9 | 964 | overall | 2026-11-09 | |
| #22 | Grok 4.7 (xHigh)xai · Proprietary | 1,631.27 | 1,647.45 | 17.03 | 7,816 | overall | 2026-11-09 | |
| #23 | MiMo-V2.6-Flashxiaomi · MIT | 1,628.69 | 1,644.36 | 15.98 | 8,576 | overall | 2026-11-09 | |
| #24 | Hy4 previewtencent · Apache 2.0 | 1,621.69 | 1,641.26 | 24.91 | 4,606 | overall | 2026-11-09 | |
| #25 | Claude Fable 5 (High)anthropic · Proprietary | 1,619.22 | 1,631.17 | 9.3 | 17,017 | overall | 2026-11-09 | |
| #26 | Qwen3.8-Flash-Nextalibaba · qwen-community-1.0 | 1,617.24 | 1,631.7 | 13.6 | 10,939 | overall | 2026-11-09 | |
| #27 | MiMo-V2.6-Proxiaomi · MIT | 1,616.11 | 1,631.71 | 15.83 | 8,402 | overall | 2026-11-09 | |
| #28 | GLM-5.3 (Max)zai · MIT | 1,615.22 | 1,629.13 | 12.59 | 12,263 | overall | 2026-11-09 | |
| #29 | DeepSeek-V4.1-Flash-20260910 (Max)deepseek · MIT | 1,609.59 | 1,631.83 | 32.18 | 3,408 | overall | 2026-11-09 | |
| #30 | Muse Spark 1.3 (xHigh)meta · Proprietary | 1,613.43 | 1,627.6 | 13.07 | 11,277 | overall | 2026-11-09 | |
| #31 | Grok 4.6 (High)xai · Proprietary | 1,610.87 | 1,624.77 | 12.57 | 12,193 | overall | 2026-11-09 | |
| #32 | GPT-5.6 Sol (xHigh, Codex Harness)openai · Proprietary | 1,609.55 | 1,621.01 | 8.53 | 18,929 | overall | 2026-11-09 | |
| #33 | GLM-5.3-Flashzai · MIT | 1,602.37 | 1,614.78 | 10.02 | 23,185 | overall | 2026-11-09 | |
| #34 | Muse Ember (Max)meta · Proprietary | 1,591.32 | 1,624.97 | 73.69 | 1,345 | overall | 2026-11-09 | |
| #35 | GLM-5.2 (Max)zai · MIT | 1,601.88 | 1,613.66 | 9.02 | 17,934 | overall | 2026-11-09 | |
| #36 | Claude Haiku 5.5 (High)anthropic · Proprietary | 1,586.42 | 1,603.83 | 19.71 | 6,174 | overall | 2026-11-09 | |
| #37 | Qwen3.8-27Balibaba · Apache 2.0 | 1,584.39 | 1,596.64 | 9.76 | 20,408 | overall | 2026-11-09 | |
| #38 | Gemini 3.7 Flash (High)google · Proprietary | 1,581.69 | 1,594.63 | 10.89 | 14,619 | overall | 2026-11-09 | |
| #39 | Gemini 3.8 Flash (High)google · Proprietary | 1,577.07 | 1,590.53 | 11.79 | 13,485 | overall | 2026-11-09 | |
| #40 | DeepSeek-V4-Flash (High)deepseek · MIT | 1,572.44 | 1,591.23 | 22.96 | 4,839 | overall | 2026-11-09 | |
| #41 | MiniMax-M3.1-Flashminimax · MiniMax Community License | 1,571.62 | 1,591.57 | 25.89 | 4,421 | overall | 2026-11-09 | |
| #42 | DeepSeek-V4-Pro (High, 20260813)deepseek · MIT | 1,571.66 | 1,590.72 | 23.63 | 4,614 | overall | 2026-11-09 | |
| #43 | GPT-6 Luna (Max)openai · Proprietary | 1,572.34 | 1,584.15 | 9.07 | 29,989 | overall | 2026-11-09 | |
| #44 | Step 5 Preview (High)stepfun · Proprietary | 1,564.21 | 1,580.72 | 17.74 | 7,267 | overall | 2026-11-09 | |
| #45 | Mistral Large 4mistral · Modified MIT | 1,545.42 | 1,579.01 | 73.41 | 1,409 | overall | 2026-11-09 | |
| #46 | Claude Opus 4.7 (High)anthropic · Proprietary | 1,551.44 | 1,563.02 | 8.73 | 17,616 | overall | 2026-11-09 | |
| #47 | Claude Opus 4.8 (High)anthropic · Proprietary | 1,549.65 | 1,561.24 | 8.74 | 18,061 | overall | 2026-11-09 | |
| #48 | Claude Opus 4.7anthropic · Proprietary | 1,549.73 | 1,561.02 | 8.3 | 17,957 | overall | 2026-11-09 | |
| #49 | Grok 4.5xai · Proprietary | 1,545.71 | 1,559.16 | 11.78 | 10,815 | overall | 2026-11-09 | |
| #50 | Claude Opus 4.6 (High)anthropic · Proprietary | 1,540.84 | 1,551.85 | 7.9 | 19,080 | overall | 2026-11-09 | |
| #51 | Muse Spark 1.1meta · Proprietary | 1,533.88 | 1,549.23 | 15.33 | 7,850 | overall | 2026-11-09 | |
| #52 | Claude Sonnet 5 (High)anthropic · Proprietary | 1,533.44 | 1,545.05 | 8.77 | 16,802 | overall | 2026-11-09 | |
| #53 | Gemini 3.6 Flash (High)google · Proprietary | 1,530.29 | 1,545.42 | 14.9 | 8,538 | overall | 2026-11-09 | |
| #54 | Claude Opus 4.6anthropic · Proprietary | 1,531.46 | 1,542.22 | 7.53 | 20,385 | overall | 2026-11-09 | |
| #55 | Claude Opus 4.8anthropic · Proprietary | 1,530.62 | 1,542.08 | 8.54 | 16,993 | overall | 2026-11-09 | |
| #56 | Muse Spark 1.2 (xHigh)meta · Proprietary | 1,519.71 | 1,547.31 | 49.58 | 2,121 | overall | 2026-11-09 | |
| #57 | Muse Spark Open (Max)meta · muse-model-license | 1,511.02 | 1,541.31 | 59.71 | 1,780 | overall | 2026-11-09 | |
| #58 | Claude Sonnet 4.6anthropic · Proprietary | 1,516.82 | 1,527.19 | 6.99 | 22,558 | overall | 2026-11-09 | |
| #59 | GPT-5.6 Terra (xHigh, Codex Harness)openai · Proprietary | 1,513.97 | 1,526.14 | 9.63 | 15,502 | overall | 2026-11-09 | |
| #60 | Seed2.1 Pro Previewbytedance · Proprietary | 1,513.59 | 1,524.91 | 8.34 | 17,367 | overall | 2026-11-09 | |
| #61 | GPT-5.6 Luna (xHigh, Codex Harness)openai · Proprietary | 1,512.54 | 1,525.46 | 10.86 | 12,231 | overall | 2026-11-09 | |
| #62 | Qwen3.7-Max (20260517)alibaba · Proprietary | 1,507.69 | 1,522.5 | 14.28 | 8,751 | overall | 2026-11-09 | |
| #63 | GPT-5.5 (xHigh, Codex Harness)openai · Proprietary | 1,505.95 | 1,517.95 | 9.37 | 15,398 | overall | 2026-11-09 | |
| #64 | Kimi K2.6moonshot · Modified MIT | 1,501.78 | 1,516.33 | 13.76 | 10,016 | overall | 2026-11-09 | |
| #65 | GLM-5.1zai · MIT | 1,501.26 | 1,514.88 | 12.09 | 10,543 | overall | 2026-11-09 | |
| #66 | Hy3tencent · Apache 2.0 | 1,497.75 | 1,513.59 | 16.32 | 7,207 | overall | 2026-11-09 | |
| #67 | Gemini 3.5 Flash (High)google · Proprietary | 1,492.23 | 1,506.86 | 13.92 | 9,167 | overall | 2026-11-09 | |
| #68 | Claude Opus 4.5 (High, 32K, 20251101)anthropic · Proprietary | 1,484.64 | 1,501.3 | 18.08 | 10,930 | overall | 2026-11-09 | |
| #69 | Gemini 3.5 Flash (Medium)google · Proprietary | 1,482.91 | 1,496.3 | 11.68 | 11,913 | overall | 2026-11-09 | |
| #70 | GPT-5.5 (High, Codex Harness)openai · Proprietary | 1,481.23 | 1,493.12 | 9.21 | 17,673 | overall | 2026-11-09 | |
| #71 | Spark-X2.6synlan · Proprietary | 1,465.63 | 1,504.01 | 95.89 | 1,189 | overall | 2026-11-09 | |
| #72 | MiniMax-M3minimax · MiniMax Community License | 1,478.76 | 1,489.58 | 7.63 | 21,077 | overall | 2026-11-09 | |
| #73 | Qwen3.6-Max-Previewalibaba · Proprietary | 1,469.54 | 1,494.68 | 41.15 | 2,677 | overall | 2026-11-09 | |
| #74 | MiMo-V2.5-Proxiaomi · MIT | 1,473.82 | 1,484.34 | 7.2 | 22,101 | overall | 2026-11-09 | |
| #75 | Kimi K2.7 Codemoonshot · Modified MIT | 1,464.42 | 1,483.31 | 23.22 | 4,891 | overall | 2026-11-09 | |
| #76 | Claude Opus 4.5 (20251101)anthropic · Proprietary | 1,461.34 | 1,475.84 | 13.68 | 13,310 | overall | 2026-11-09 | |
| #77 | DeepSeek-V4-Pro Preview (High)deepseek · MIT | 1,456.91 | 1,469.62 | 10.51 | 12,958 | overall | 2026-11-09 | |
| #78 | GPT-5.4 (High, Codex Harness)openai · Proprietary | 1,442.92 | 1,480.74 | 93.09 | 1,340 | overall | 2026-11-09 | |
| #79 | Qwen3.6-Plusalibaba · Proprietary | 1,455.79 | 1,467.01 | 8.2 | 18,273 | overall | 2026-11-09 | |
| #80 | GPT-5.5 (Codex Harness)openai · Proprietary | 1,449.46 | 1,461.29 | 9.11 | 15,634 | overall | 2026-11-09 | |
| #81 | DeepSeek-V4-Prodeepseek · MIT | 1,442 | 1,454.73 | 10.55 | 13,779 | overall | 2026-11-09 | |
| #82 | Gemini 3.1 Pro Previewgoogle · Proprietary | 1,441.29 | 1,451.47 | 6.75 | 23,865 | overall | 2026-11-09 | |
| #83 | GPT-5.4 (Medium, Codex Harness)openai · Proprietary | 1,423.64 | 1,460.37 | 87.78 | 1,363 | overall | 2026-11-09 | |
| #84 | Gemini 3 Progoogle · Proprietary | 1,432.16 | 1,449.13 | 18.74 | 13,937 | overall | 2026-11-09 | |
| #85 | Gemini 3 Flashgoogle · Proprietary | 1,431.01 | 1,448.32 | 19.49 | 10,793 | overall | 2026-11-09 | |
| #86 | Beam 1.0 501B A23Breflection · Apache 2.0 | 1,424.19 | 1,451.47 | 48.42 | 2,521 | overall | 2026-11-09 | |
| #87 | MiMo-V2.5xiaomi · MIT | 1,431.77 | 1,443.88 | 9.54 | 14,646 | overall | 2026-11-09 | |
| #88 | Kimi K2.5 (Thinking)moonshot · Modified MIT | 1,430.96 | 1,441.45 | 7.16 | 19,815 | overall | 2026-11-09 | |
| #89 | Gemini 3.5 Flash-Litegoogle · Proprietary | 1,394.03 | 1,476.43 | 441.8 | 233 | overall | 2026-11-09 | |
| #90 | GLM-5zai · MIT | 1,427.01 | 1,443.4 | 17.48 | 7,418 | overall | 2026-11-09 | |
| #91 | GLM-4.7zai · MIT | 1,422.86 | 1,446.96 | 37.8 | 3,975 | overall | 2026-11-09 | |
| #92 | MiMo-V2-Proxiaomi · Proprietary | 1,425.95 | 1,442.31 | 17.42 | 7,194 | overall | 2026-11-09 | |
| #93 | DeepSeek-V4-Flash Preview (High)deepseek · MIT | 1,423.87 | 1,437.86 | 12.73 | 9,986 | overall | 2026-11-09 | |
| #94 | GPT-5.2openai · Proprietary | 1,396.88 | 1,442.49 | 135.4 | 1,065 | overall | 2026-11-09 | |
| #95 | GPT-5 (Medium)openai · Proprietary | 1,401.04 | 1,432.37 | 63.86 | 3,036 | overall | 2026-11-09 | |
| #96 | Inklingthinky · Apache 2.0 | 1,408.17 | 1,420.14 | 9.31 | 17,206 | overall | 2026-11-09 | |
| #97 | Inkling-Smallthinky · Apache 2.0 | 1,398.17 | 1,416.49 | 21.85 | 5,228 | overall | 2026-11-09 | |
| #98 | GPT-5.3-Codex (Codex Harness)openai · Proprietary | 1,392.02 | 1,419.46 | 49.01 | 2,630 | overall | 2026-11-09 | |
| #99 | Kimi K2.5 (Instant)moonshot · Modified MIT | 1,393.39 | 1,417.61 | 38.18 | 3,244 | overall | 2026-11-09 | |
| #100 | Qwen3.5-397B-A17Balibaba · Apache 2.0 | 1,394.95 | 1,405.27 | 6.93 | 20,871 | overall | 2026-11-09 | |
| #101 | GPT-5.4openai · Proprietary | 1,389.63 | 1,409.3 | 25.17 | 4,666 | overall | 2026-11-09 | |
| #102 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)nvidia · OpenMDW-1.1 | 1,388.24 | 1,410.6 | 32.53 | 4,660 | overall | 2026-11-09 | |
| #103 | GLM-5V-Turbozai · Proprietary | 1,386.03 | 1,412.6 | 45.95 | 2,375 | overall | 2026-11-09 | |
| #104 | MiniMax-M2.7minimax · Modified MIT | 1,391.88 | 1,403.54 | 8.85 | 15,748 | overall | 2026-11-09 | |
| #105 | GPT-5.4 mini (High)openai · Proprietary | 1,390.96 | 1,404.08 | 11.2 | 11,823 | overall | 2026-11-09 | |
| #106 | GPT-5.1 (Medium)openai · Proprietary | 1,385.51 | 1,408.71 | 35.04 | 4,859 | overall | 2026-11-09 | |
| #107 | Claude Sonnet 4.5 (High, 32K, 20250929)anthropic · Proprietary | 1,385.62 | 1,401.14 | 15.68 | 13,102 | overall | 2026-11-09 | |
| #108 | Claude Opus 4.1 (20250805)anthropic · Proprietary | 1,378.64 | 1,400.21 | 30.27 | 6,816 | overall | 2026-11-09 | |
| #109 | MiniMax-M2.5minimax · Modified MIT | 1,377.16 | 1,393.9 | 18.25 | 7,576 | overall | 2026-11-09 | |
| #110 | Claude Sonnet 4.5 (20250929)anthropic · Proprietary | 1,377.23 | 1,391.89 | 13.98 | 15,553 | overall | 2026-11-09 | |
| #111 | Gemini 3 Flash (Minimal)google · Proprietary | 1,379.06 | 1,389.58 | 7.19 | 21,966 | overall | 2026-11-09 | |
| #112 | MiniMax-M2.1 Previewminimax · MIT | 1,374.73 | 1,393.9 | 23.92 | 7,581 | overall | 2026-11-09 | |
| #113 | Grok 4.20 Beta (Reasoning, 0309)xai · Proprietary | 1,369.41 | 1,381.63 | 9.72 | 15,278 | overall | 2026-11-09 | |
| #114 | GPT-5.3-Codex (Codex Harness)openai · Proprietary | 1,359.56 | 1,382.27 | 33.56 | 3,631 | overall | 2026-11-09 | |
| #115 | Gemma 4 31Bgoogle · Apache 2.0 | 1,363.14 | 1,375.22 | 9.49 | 16,251 | overall | 2026-11-09 | |
| #116 | Solar Pro 4upstage · Proprietary | 1,358.5 | 1,379.46 | 28.59 | 4,503 | overall | 2026-11-09 | |
| #117 | DeepSeek-V3.2 (Thinking)deepseek · MIT | 1,353.37 | 1,371.66 | 21.77 | 6,858 | overall | 2026-11-09 | |
| #118 | Qwen3.5-122B-A10Balibaba · Apache 2.0 | 1,352.4 | 1,367.78 | 15.4 | 8,190 | overall | 2026-11-09 | |
| #119 | Gemma 4 26B A4Bgoogle · Apache 2.0 | 1,341.46 | 1,375.37 | 74.8 | 1,492 | overall | 2026-11-09 | |
| #120 | Hy3 previewtencent · tencent-hunyuan-community | 1,340.73 | 1,375.74 | 79.76 | 1,424 | overall | 2026-11-09 | |
| #121 | Grok 4.3xai · Proprietary | 1,351.17 | 1,363.74 | 10.28 | 14,694 | overall | 2026-11-09 | |
| #122 | Qwen3.5-27Balibaba · Apache 2.0 | 1,349.02 | 1,365.16 | 16.95 | 7,738 | overall | 2026-11-09 | |
| #123 | Muse Glimmermeta · Apache-2.0 | 1,338.88 | 1,370.23 | 63.96 | 1,634 | overall | 2026-11-09 | |
| #124 | Laguna M.1poolside · Apache 2.0 | 1,340.08 | 1,358.74 | 22.67 | 5,319 | overall | 2026-11-09 | |
| #125 | GPT-5.1openai · Proprietary | 1,333.61 | 1,350.69 | 18.97 | 10,600 | overall | 2026-11-09 | |
| #126 | GLM-4.6zai · MIT | 1,331.24 | 1,352.96 | 30.71 | 6,680 | overall | 2026-11-09 | |
| #127 | GPT-5.2-Codexopenai · Proprietary | 1,329.34 | 1,347.29 | 20.98 | 6,669 | overall | 2026-11-09 | |
| #128 | GPT-5.1-Codexopenai · Proprietary | 1,325.99 | 1,349.31 | 35.39 | 4,941 | overall | 2026-11-09 | |
| #129 | MiMo-V2-Flash (Non-Thinking)xiaomi · MIT | 1,321.34 | 1,340.93 | 24.98 | 5,647 | overall | 2026-11-09 | |
| #130 | Claude Haiku 4.5 (20251001)anthropic · Proprietary | 1,325.11 | 1,335.05 | 6.43 | 28,706 | overall | 2026-11-09 | |
| #131 | DeepSeek-V3.2deepseek · MIT | 1,317.67 | 1,333.69 | 16.7 | 9,302 | overall | 2026-11-09 | |
| #132 | Kimi K2 Thinking Turbomoonshot · Modified MIT | 1,313.86 | 1,328.86 | 14.64 | 12,962 | overall | 2026-11-09 | |
| #133 | Laguna XS.2poolside · Apache 2.0 | 1,292.44 | 1,313.9 | 29.96 | 4,166 | overall | 2026-11-09 | |
| #134 | MiniMax-M2minimax · Apache 2.0 | 1,287.93 | 1,309.94 | 31.51 | 6,606 | overall | 2026-11-09 | |
| #135 | MiMo-V2-Flash (Thinking)xiaomi · MIT | 1,275.35 | 1,308.35 | 70.89 | 1,668 | overall | 2026-11-09 | |
| #136 | Qwen3-Coder-480B-A35B-Instructalibaba · Apache 2.0 | 1,266.71 | 1,282.41 | 16.03 | 12,576 | overall | 2026-11-09 | |
| #137 | DeepSeek-V3.2-Expdeepseek · MIT | 1,256.66 | 1,283.88 | 48.21 | 3,902 | overall | 2026-11-09 | |
| #138 | Mistral Medium 3.5mistral · Modified MIT | 1,247.65 | 1,277.36 | 57.45 | 2,287 | overall | 2026-11-09 | |
| #139 | KAT-Coder-Pro V1Proprietary | 1,237.66 | 1,276.79 | 99.6 | 1,493 | overall | 2026-11-09 | |
| #140 | Gemini 3.1 Flash-Lite Previewgoogle · Proprietary | 1,248.72 | 1,262.47 | 12.3 | 14,259 | overall | 2026-11-09 | |
| #141 | Qwen3.5-35B-A3Balibaba · Apache 2.0 | 1,235.22 | 1,270.18 | 79.52 | 1,613 | overall | 2026-11-09 | |
| #142 | GPT-5.1-Codex-Miniopenai · Proprietary | 1,222.06 | 1,266.74 | 129.91 | 1,123 | overall | 2026-11-09 | |
| #143 | Qwen3.5-Flashalibaba · Proprietary | 1,223.8 | 1,263.2 | 101 | 1,322 | overall | 2026-11-09 | |
| #144 | Grok 4.1 Fast (Reasoning)xai · Proprietary | 1,230.56 | 1,252.87 | 32.41 | 5,437 | overall | 2026-11-09 | |
| #145 | Trinity Large ThinkingApache 2.0 | 1,216.32 | 1,256.44 | 104.78 | 1,350 | overall | 2026-11-09 | |
| #146 | Mistral Large 3mistral · Apache 2.0 | 1,207.24 | 1,259.04 | 174.59 | 816 | overall | 2026-11-09 | |
| #147 | Gemini 2.5 Progoogle · Proprietary | 1,211.53 | 1,243.39 | 66.06 | 2,676 | overall | 2026-11-09 | |
| #148 | Grok 4.1 (Thinking)xai · Proprietary | 1,191.66 | 1,241.53 | 161.84 | 955 | overall | 2026-11-09 | |
| #149 | Devstral 2mistral · Modified MIT | 1,174.67 | 1,216.29 | 112.73 | 1,171 | overall | 2026-11-09 | |
| #150 | Granite 4.1 8Bibm · Apache 2.0 | 1,174 | 1,211.17 | 89.9 | 1,748 | overall | 2026-11-09 | |
| #151 | Mercury 2inception-ai · Proprietary | 1,146.33 | 1,195.69 | 158.6 | 940 | overall | 2026-11-09 | |
| #152 | Grok Code Fast 1xai · Proprietary | 1,135.16 | 1,192.56 | 214.45 | 770 | overall | 2026-11-09 | |
| #153 | Grok 4 Fast (Reasoning)xai · Proprietary | 1,135.22 | 1,190.71 | 200.34 | 715 | overall | 2026-11-09 | |
| #154 | Devstral Medium (2507)mistral · Proprietary | 1,041.01 | 1,102.55 | 246.44 | 833 | overall | 2026-11-09 |
用113个从零编写的长程真实软件工程任务测评编程Agent,同时刻画准确率(Pass@1)与真实花费/步数的性价比曲线,避免题目训练污染。
在新标签页打开独立跑分,不依赖厂商自报数据;综合 Intelligence Index 覆盖推理/编程/科学/知识,并联动速度、价格、上下文窗口等工程指标。
在新标签页打开每月更新全新题目防污染,全自动客观判分(唯一正确答案),聚焦数学/代码/推理/数据分析,学术圈公认权威。
在新标签页打开中文原创题库,深度贴合中文语境与政策文化,覆盖数学推理、代码、幻觉控制、智能体规划六大能力。
在新标签页打开上海AI实验室出品,20余家企业共建,五大能力维度、70+ 数据集,中英文模型均覆盖,并含真人投票的 Compass Arena。
在新标签页打开本页对战数据的官方原始来源,累计投票超过700万次,覆盖370+模型,逐票更新,界面本身也值得参考。
在新标签页打开