作舟知识辅助
AI 榜单
从智能体、聊天、编程、图像和视频五类榜单出发,找到值得亲自试用的模型。排名只是参考,最终选择仍取决于你的任务与预算。
Agent
智能体
来自真实智能体会话的综合榜,结合任务成功、用户满意度和可引导性等信号。净提升及置信区间以百分比表示。
每日同步正常
左右滚动查看全部字段。数值保留两位小数,悬停可查看原始精度;原始分类和数据日期按来源保留。
| 排名 | 模型 | 开发方 | 模型许可证 | 净提升 | 95% 置信下界 | 95% 置信上界 | 观测数 | 会话数 | 原始分类 | 数据发布日期 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 (Max) | anthropic | Proprietary | 20.35% | 17.79% | 22.91% | 1,562,084 | 11,182 | overall | 2026-10-19 |
| 2 | GPT-6 Astra (Max) | openai | Proprietary | 15.15% | 12.91% | 17.39% | 1,055,314 | 13,792 | overall | 2026-10-19 |
| 3 | Claude Fable 5.1 (Max) | anthropic | Proprietary | 14.33% | 12.34% | 16.31% | 1,962,103 | 16,607 | overall | 2026-10-19 |
| 4 | GPT-6.1 Sol (Max) | openai | Proprietary | 13.72% | 11.95% | 15.49% | 1,187,713 | 14,746 | overall | 2026-10-19 |
| 5 | Claude Sonnet 5.5 (Max) | anthropic | Proprietary | 12.75% | 9.71% | 15.79% | 997,013 | 7,769 | overall | 2026-10-19 |
| 6 | Gemini 4 Argon (High) | Proprietary | 12.12% | 11.01% | 13.24% | 3,577,959 | 35,505 | overall | 2026-10-19 | |
| 7 | GPT-6 Sol (Max) | openai | Proprietary | 9.33% | 7.50% | 11.16% | 1,823,820 | 9,246 | overall | 2026-10-19 |
| 8 | Claude Fable 5 (High) | anthropic | Proprietary | 8.37% | 6.92% | 9.82% | 2,589,869 | 43,116 | overall | 2026-10-19 |
| 9 | Claude Opus 5 (Max) | anthropic | Proprietary | 8.08% | 6.41% | 9.75% | 3,216,618 | 24,050 | overall | 2026-10-19 |
| 10 | GPT-5.6 Sol (xHigh) | openai | Proprietary | 6.59% | 5.21% | 7.98% | 4,546,761 | 37,049 | overall | 2026-10-19 |
| 11 | Claude Sonnet 5 (High) | anthropic | Proprietary | 4.06% | 2.39% | 5.73% | 3,648,024 | 33,569 | overall | 2026-10-19 |
| 12 | Muse Wonder (Max) | meta | Proprietary | 3.77% | 2.15% | 5.40% | 1,179,447 | 9,689 | overall | 2026-10-19 |
| 13 | MiMo-V2.6-Pro | xiaomi | MIT | 3.74% | 2.82% | 4.66% | 3,458,520 | 36,102 | overall | 2026-10-19 |
| 14 | Gemini 3.8 Flash (High) | Proprietary | 3.69% | 2.58% | 4.80% | 5,460,958 | 38,204 | overall | 2026-10-19 | |
| 15 | Muse Spark 1.3 (Max) | meta | Proprietary | 3.33% | 2.49% | 4.16% | 7,549,812 | 74,953 | overall | 2026-10-19 |
| 16 | Grok 4.7 (xHigh) | xai | Proprietary | 3.30% | 1.70% | 4.90% | 2,623,682 | 15,665 | overall | 2026-10-19 |
| 17 | GPT-5.5 (xHigh) | openai | Proprietary | 2.95% | 1.68% | 4.21% | 2,722,987 | 54,351 | overall | 2026-10-19 |
| 18 | GPT-6 Luna (Max) | openai | Proprietary | 2.86% | 2.43% | 3.28% | 27,301,901 | 182,171 | overall | 2026-10-19 |
| 19 | DeepSeek-V4.1-Flash-20260910 (Max) | deepseek | MIT | 2.56% | 2.21% | 2.91% | 40,819,636 | 255,742 | overall | 2026-10-19 |
| 20 | MiMo-V2.6-Flash | xiaomi | MIT | 2.54% | 1.15% | 3.94% | 2,311,004 | 20,935 | overall | 2026-10-19 |
| 21 | Hy4 preview | tencent | Apache 2.0 | 2.47% | 1.81% | 3.12% | 9,052,741 | 88,908 | overall | 2026-10-19 |
| 22 | GLM-5.2 (Max) | zai | MIT | 2.13% | 1.29% | 2.97% | 7,011,932 | 88,576 | overall | 2026-10-19 |
| 23 | Kimi K3 (Max) | moonshot | Kimi K3 license | 1.89% | 1.14% | 2.64% | 12,609,714 | 136,966 | overall | 2026-10-19 |
| 24 | Qwen3.8-Max | alibaba | Proprietary | 1.63% | 1.02% | 2.24% | 6,745,555 | 71,737 | overall | 2026-10-19 |
| 25 | Claude Haiku 5.5 (High) | anthropic | Proprietary | 1.49% | 0.49% | 2.48% | 2,860,691 | 26,311 | overall | 2026-10-19 |
| 26 | GLM-5.3 (Max) | zai | MIT | 1.31% | 0.47% | 2.16% | 9,469,189 | 87,071 | overall | 2026-10-19 |
| 27 | GPT-5.4 (High) | openai | Proprietary | 1.25% | 0.17% | 2.33% | 3,887,759 | 81,847 | overall | 2026-10-19 |
| 28 | Grok 4.6 (xHigh) | xai | Proprietary | 0.91% | -0.22% | 2.04% | 3,389,136 | 29,247 | overall | 2026-10-19 |
| 29 | GPT-5.6 Terra (xHigh) | openai | Proprietary | 0.28% | -1.01% | 1.58% | 1,730,312 | 24,677 | overall | 2026-10-19 |
| 30 | DeepSeek-V4-Pro (High, 20260813) | deepseek | MIT | -0.04% | -1.81% | 1.73% | 847,761 | 7,070 | overall | 2026-10-19 |
| 31 | GPT-5.5 | openai | Proprietary | -0.40% | -1.64% | 0.83% | 2,206,363 | 82,635 | overall | 2026-10-19 |
| 32 | Gemini 3.7 Flash (High) | Proprietary | -1.06% | -2.02% | -0.10% | 5,011,708 | 55,513 | overall | 2026-10-19 | |
| 33 | GLM-5.3-Flash | zai | MIT | -1.54% | -2.23% | -0.85% | 11,568,139 | 88,684 | overall | 2026-10-19 |
| 34 | Step 5 Preview | stepfun | Proprietary | -1.84% | -3.32% | -0.36% | 2,121,873 | 14,526 | overall | 2026-10-19 |
| 35 | GPT-5.6 Luna (xHigh) | openai | Proprietary | -2.23% | -2.98% | -1.48% | 4,091,361 | 43,317 | overall | 2026-10-19 |
| 36 | Qwen3.8-Flash-Next | alibaba | qwen-community-1.0 | -3.03% | -3.71% | -2.36% | 14,385,676 | 101,114 | overall | 2026-10-19 |
| 37 | Qwen3.8-27B | alibaba | Apache 2.0 | -3.80% | -4.59% | -3.01% | 7,499,760 | 61,114 | overall | 2026-10-19 |
| 38 | Gemini 3.1 Pro Preview | Proprietary | -3.90% | -5.24% | -2.55% | 2,849,963 | 87,422 | overall | 2026-10-19 | |
| 39 | Mistral Large 4 | mistral | Proprietary | -4.46% | -6.38% | -2.53% | 1,008,534 | 14,035 | overall | 2026-10-19 |
| 40 | Beam 1.0 501B A23B | reflection | Apache 2.0 | -4.60% | -6.83% | -2.38% | 650,698 | 11,972 | overall | 2026-10-19 |
| 41 | GPT-6 Luna (Low) | openai | Proprietary | -4.80% | -6.11% | -3.49% | 745,773 | 12,307 | overall | 2026-10-19 |
| 42 | Hy3 | tencent | Apache 2.0 | -6.27% | -7.75% | -4.79% | 1,167,533 | 31,609 | overall | 2026-10-19 |
| 43 | Qwen3.7-Max | alibaba | Proprietary | -6.34% | -7.50% | -5.17% | 2,153,897 | 42,377 | overall | 2026-10-19 |
| 44 | Muse Spark 1.1 | meta | Proprietary | -6.57% | -7.23% | -5.91% | 7,861,186 | 146,559 | overall | 2026-10-19 |
| 45 | Qwen3.7-Plus | alibaba | Proprietary | -6.69% | -8.25% | -5.12% | 1,184,324 | 23,630 | overall | 2026-10-19 |
| 46 | Muse Spark 1.2 (xHigh) | meta | Proprietary | -6.73% | -7.57% | -5.90% | 3,351,482 | 52,015 | overall | 2026-10-19 |
| 47 | MiMo-V2.5-Pro | xiaomi | MIT | -8.21% | -9.24% | -7.18% | 2,444,816 | 49,148 | overall | 2026-10-19 |
| 48 | MiniMax-M3 | minimax | MiniMax Community License | -9.49% | -10.73% | -8.25% | 3,371,405 | 43,277 | overall | 2026-10-19 |
| 49 | Nemotron 3 Ultra | nvidia | OpenMDW-1.1 | -10.83% | -15.51% | -6.14% | 177,688 | 13,043 | overall | 2026-10-19 |
| 50 | Mistral Medium 3.5 | mistral | Modified MIT | -11.83% | -13.57% | -10.09% | 556,555 | 13,095 | overall | 2026-10-19 |
| 51 | Inkling | thinky | Apache 2.0 | -14.67% | -16.10% | -13.24% | 1,677,213 | 43,692 | overall | 2026-10-19 |
| 52 | Inkling-Small | thinky | Apache 2.0 | -15.68% | -17.50% | -13.86% | 512,901 | 14,197 | overall | 2026-10-19 |
| 53 | Solar Pro 4 | upstage | Proprietary | -19.69% | -21.98% | -17.39% | 471,433 | 10,406 | overall | 2026-10-19 |