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。
Chat
聊天
Arena 默认文本综合榜,采用风格控制后的用户偏好评分,帮助比较通用聊天和文本回答表现。
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 | Arena 分数 | 置信下界 | 置信上界 | 评分方差 | 对战次数 | 原始分类 | 数据发布日期 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argon (High) | Proprietary | 1,523.9 | 1,515.09 | 1,532.72 | 20.24 | 4,904 | overall | 2026-10-15 | |
| 2 | Muse Wonder (Max) | meta | Proprietary | 1,508.88 | 1,498.31 | 1,519.45 | 29.09 | 3,247 | overall | 2026-10-15 |
| 3 | Claude Opus 5.5 (High) | anthropic | Proprietary | 1,508.17 | 1,500.84 | 1,515.5 | 13.99 | 8,300 | overall | 2026-10-15 |
| 4 | Claude Fable 5 (High) | anthropic | Proprietary | 1,503.6 | 1,499.31 | 1,507.89 | 4.79 | 42,004 | overall | 2026-10-15 |
| 5 | Claude Opus 4.6 (High) | anthropic | Proprietary | 1,503.05 | 1,499.71 | 1,506.39 | 2.91 | 81,394 | overall | 2026-10-15 |
| 6 | Claude Fable 5.1 (Max) | anthropic | Proprietary | 1,502.29 | 1,496.24 | 1,508.33 | 9.52 | 14,590 | overall | 2026-10-15 |
| 7 | Claude Opus 4.7 (High) | anthropic | Proprietary | 1,501.06 | 1,497.3 | 1,504.82 | 3.68 | 65,116 | overall | 2026-10-15 |
| 8 | Claude Opus 4.6 | anthropic | Proprietary | 1,497.21 | 1,493.95 | 1,500.46 | 2.76 | 86,071 | overall | 2026-10-15 |
| 9 | Gemini 3.8 Flash (High) | Proprietary | 1,496.38 | 1,491.75 | 1,501.01 | 5.58 | 37,454 | overall | 2026-10-15 | |
| 10 | Muse Spark 1.3 (Max) | meta | Proprietary | 1,494.41 | 1,488.82 | 1,500.01 | 8.15 | 17,561 | overall | 2026-10-15 |
| 11 | Claude Opus 4.7 | anthropic | Proprietary | 1,493.59 | 1,489.83 | 1,497.35 | 3.68 | 66,216 | overall | 2026-10-15 |
| 12 | Muse Spark 1.2 (xHigh) | meta | Proprietary | 1,493.06 | 1,485.9 | 1,500.22 | 13.36 | 8,248 | overall | 2026-10-15 |
| 13 | Muse Spark 1.1 | meta | Proprietary | 1,490.18 | 1,485.93 | 1,494.42 | 4.69 | 42,810 | overall | 2026-10-15 |
| 14 | Claude Opus 5 (Max) | anthropic | Proprietary | 1,489.99 | 1,485.16 | 1,494.82 | 6.07 | 32,495 | overall | 2026-10-15 |
| 15 | Claude Opus 5 (High) | anthropic | Proprietary | 1,489.91 | 1,486.24 | 1,493.58 | 3.51 | 66,824 | overall | 2026-10-15 |
| 16 | Muse Spark | meta | Proprietary | 1,489.04 | 1,483.31 | 1,494.77 | 8.55 | 14,150 | overall | 2026-10-15 |
| 17 | Gemini 3.7 Flash (High) | Proprietary | 1,487.59 | 1,482.71 | 1,492.47 | 6.2 | 27,232 | overall | 2026-10-15 | |
| 18 | Kimi K3 (Max) | moonshot | Kimi K3 license | 1,487.07 | 1,482.49 | 1,491.66 | 5.47 | 32,795 | overall | 2026-10-15 |
| 19 | Gemini 3.1 Pro Preview | Proprietary | 1,486.91 | 1,483.95 | 1,489.87 | 2.28 | 127,649 | overall | 2026-10-15 | |
| 20 | Gemini 3 Pro | Proprietary | 1,485.46 | 1,481.67 | 1,489.25 | 3.74 | 42,072 | overall | 2026-10-15 | |
| 21 | gpt-5.6-sol-xhigh | openai | Proprietary | 1,484.74 | 1,480.4 | 1,489.08 | 4.9 | 40,442 | overall | 2026-10-15 |
| 22 | gpt-6.1-sol-max | openai | Proprietary | 1,483.85 | 1,475.71 | 1,491.98 | 17.23 | 6,320 | overall | 2026-10-15 |
| 23 | gemini-3.6-flash-high | Proprietary | 1,482.7 | 1,478.28 | 1,487.13 | 5.09 | 36,122 | overall | 2026-10-15 | |
| 24 | qwen3.8-max | alibaba | Proprietary | 1,482.35 | 1,477.38 | 1,487.33 | 6.45 | 27,766 | overall | 2026-10-15 |
| 25 | claude-opus-4-8-high | anthropic | Proprietary | 1,481.5 | 1,477.63 | 1,485.36 | 3.89 | 64,100 | overall | 2026-10-15 |
| 26 | gpt-5.5-high | openai | Proprietary | 1,481.44 | 1,477.72 | 1,485.16 | 3.61 | 69,401 | overall | 2026-10-15 |
| 27 | mimo-v2.6-pro | xiaomi | MIT | 1,479.19 | 1,469.83 | 1,488.56 | 22.84 | 4,075 | overall | 2026-10-15 |
| 28 | glm-5.3-max | zai | MIT | 1,478.54 | 1,473.31 | 1,483.77 | 7.12 | 22,543 | overall | 2026-10-15 |
| 29 | gemini-3.5-flash-high | Proprietary | 1,477.48 | 1,473.58 | 1,481.37 | 3.94 | 48,486 | overall | 2026-10-15 | |
| 30 | gpt-6-astra-max | openai | Proprietary | 1,477.32 | 1,471.02 | 1,483.62 | 10.33 | 12,163 | overall | 2026-10-15 |
| 31 | gpt-5.5 | openai | Proprietary | 1,476.67 | 1,472.98 | 1,480.36 | 3.54 | 70,985 | overall | 2026-10-15 |
| 32 | glm-5.2-max | zai | MIT | 1,475.88 | 1,471.7 | 1,480.06 | 4.55 | 49,820 | overall | 2026-10-15 |
| 33 | gemini-3.5-flash-medium | Proprietary | 1,475.67 | 1,471.52 | 1,479.81 | 4.47 | 47,292 | overall | 2026-10-15 | |
| 34 | gpt-5.2-chat-latest-20260210 | openai | Proprietary | 1,475.64 | 1,471.6 | 1,479.68 | 4.25 | 35,999 | overall | 2026-10-15 |
| 35 | gpt-5.4-high | openai | Proprietary | 1,475.12 | 1,471.36 | 1,478.89 | 3.69 | 64,776 | overall | 2026-10-15 |
| 36 | claude-sonnet-5.5-xhigh | anthropic | Proprietary | 1,474.94 | 1,467.2 | 1,482.68 | 15.6 | 6,831 | overall | 2026-10-15 |
| 37 | grok-4.20-beta1 | xai | Proprietary | 1,474.73 | 1,470.16 | 1,479.3 | 5.44 | 27,908 | overall | 2026-10-15 |
| 38 | deepseek-v4.1-flash-max | deepseek | MIT | 1,474.63 | 1,468.41 | 1,480.86 | 10.08 | 12,793 | overall | 2026-10-15 |
| 39 | qwen3.7-max-preview | alibaba | Proprietary | 1,474.35 | 1,464.62 | 1,484.07 | 24.63 | 3,934 | overall | 2026-10-15 |
| 40 | claude-opus-4-8 | anthropic | Proprietary | 1,474.32 | 1,470.47 | 1,478.17 | 3.85 | 65,172 | overall | 2026-10-15 |
| 41 | glm-5.3-flash | zai | MIT | 1,473.84 | 1,468.97 | 1,478.71 | 6.18 | 32,925 | overall | 2026-10-15 |
| 42 | claude-opus-4-5-20251101-high-32k | anthropic | Proprietary | 1,473.66 | 1,469.83 | 1,477.5 | 3.82 | 37,509 | overall | 2026-10-15 |
| 43 | gpt-5.5-instant | openai | Proprietary | 1,473.11 | 1,468.11 | 1,478.11 | 6.51 | 27,158 | overall | 2026-10-15 |
| 44 | gemini-3-flash | Proprietary | 1,472.8 | 1,468.45 | 1,477.14 | 4.91 | 31,325 | overall | 2026-10-15 | |
| 45 | claude-sonnet-4-6 | anthropic | Proprietary | 1,472.1 | 1,468.59 | 1,475.62 | 3.21 | 70,931 | overall | 2026-10-15 |
| 46 | grok-4.20-beta-0309-reasoning | xai | Proprietary | 1,471.55 | 1,467.88 | 1,475.22 | 3.51 | 66,436 | overall | 2026-10-15 |
| 47 | grok-4.20-multi-agent-beta-0309 | xai | Proprietary | 1,470.31 | 1,466.61 | 1,474.01 | 3.56 | 64,946 | overall | 2026-10-15 |
| 48 | claude-opus-4-5-20251101 | anthropic | Proprietary | 1,469.63 | 1,466.5 | 1,472.76 | 2.55 | 72,899 | overall | 2026-10-15 |
| 49 | mimo-v2.5-pro | xiaomi | MIT | 1,467.26 | 1,463.72 | 1,470.8 | 3.26 | 75,721 | overall | 2026-10-15 |
| 50 | ernie-5.1 | baidu | Proprietary | 1,467.01 | 1,462.47 | 1,471.55 | 5.37 | 39,468 | overall | 2026-10-15 |
| 51 | gpt-5.6-terra-xhigh | openai | Proprietary | 1,466.43 | 1,462.17 | 1,470.69 | 4.72 | 43,271 | overall | 2026-10-15 |
| 52 | grok-4.5 | xai | Proprietary | 1,465.71 | 1,461.37 | 1,470.05 | 4.9 | 39,896 | overall | 2026-10-15 |
| 53 | grok-4.1-thinking | xai | Proprietary | 1,465.18 | 1,462 | 1,468.37 | 2.63 | 66,509 | overall | 2026-10-15 |
| 54 | glm-5.1 | zai | MIT | 1,465.17 | 1,461.58 | 1,468.75 | 3.35 | 62,889 | overall | 2026-10-15 |
| 55 | gpt-5.4 | openai | Proprietary | 1,464.74 | 1,461.05 | 1,468.43 | 3.54 | 67,870 | overall | 2026-10-15 |
| 56 | deepseek-v4-pro-high-20260813 | deepseek | MIT | 1,464.58 | 1,457.97 | 1,471.19 | 11.38 | 9,800 | overall | 2026-10-15 |
| 57 | qwen3.5-max-preview | alibaba | Proprietary | 1,464.36 | 1,459.48 | 1,469.24 | 6.21 | 22,600 | overall | 2026-10-15 |
| 58 | claude-sonnet-5-high | anthropic | Proprietary | 1,461.07 | 1,456.95 | 1,465.2 | 4.43 | 49,595 | overall | 2026-10-15 |
| 59 | kimi-k2.6 | moonshot | Modified MIT | 1,460.83 | 1,456.4 | 1,465.26 | 5.1 | 40,013 | overall | 2026-10-15 |
| 60 | qwen3.6-max-preview | alibaba | Proprietary | 1,460.58 | 1,452.39 | 1,468.77 | 17.46 | 5,426 | overall | 2026-10-15 |
| 61 | gpt-6-sol-max | openai | Proprietary | 1,459.02 | 1,452.68 | 1,465.37 | 10.49 | 12,466 | overall | 2026-10-15 |
| 62 | grok-4.1 | xai | Proprietary | 1,459.02 | 1,455.86 | 1,462.19 | 2.61 | 68,648 | overall | 2026-10-15 |
| 63 | gemini-3-flash (thinking-minimal) | Proprietary | 1,458.29 | 1,455.27 | 1,461.31 | 2.38 | 91,039 | overall | 2026-10-15 | |
| 64 | deepseek-v4-pro | deepseek | MIT | 1,457.79 | 1,453.89 | 1,461.69 | 3.95 | 57,594 | overall | 2026-10-15 |
| 65 | glm-5 | zai | MIT | 1,457.27 | 1,453.02 | 1,461.51 | 4.69 | 29,272 | overall | 2026-10-15 |
| 66 | claude-sonnet-4-5-20250929-high-32k | anthropic | Proprietary | 1,456.4 | 1,453.64 | 1,459.16 | 1.98 | 83,940 | overall | 2026-10-15 |
| 67 | dola-seed-2.0-pro | bytedance | Proprietary | 1,456.17 | 1,452.78 | 1,459.57 | 3 | 79,140 | overall | 2026-10-15 |
| 68 | gpt-5.1-high | openai | Proprietary | 1,455.72 | 1,452.01 | 1,459.42 | 3.58 | 41,651 | overall | 2026-10-15 |
| 69 | qwen3.7-plus | alibaba | Proprietary | 1,455.68 | 1,451.35 | 1,460.02 | 4.89 | 42,075 | overall | 2026-10-15 |
| 70 | claude-sonnet-4-5-20250929 | anthropic | Proprietary | 1,454.93 | 1,452.07 | 1,457.79 | 2.13 | 82,688 | overall | 2026-10-15 |
| 71 | deepseek-v4-pro-high-preview | deepseek | MIT | 1,454.86 | 1,450.9 | 1,458.83 | 4.09 | 54,902 | overall | 2026-10-15 |
| 72 | gemini-3.5-flash-lite | Proprietary | 1,454.76 | 1,450.29 | 1,459.23 | 5.21 | 35,851 | overall | 2026-10-15 | |
| 73 | grok-4.6-high | xai | Proprietary | 1,454.49 | 1,449.65 | 1,459.32 | 6.08 | 28,965 | overall | 2026-10-15 |
| 74 | hy3 | tencent | Apache 2.0 | 1,453.19 | 1,447.04 | 1,459.34 | 9.84 | 12,208 | overall | 2026-10-15 |
| 75 | gpt-5.6-luna-xhigh | openai | Proprietary | 1,452.57 | 1,448.26 | 1,456.87 | 4.82 | 41,109 | overall | 2026-10-15 |
| 76 | gemma-4-31b | Apache 2.0 | 1,452.16 | 1,444.73 | 1,459.6 | 14.4 | 6,147 | overall | 2026-10-15 | |
| 77 | mimo-v2.6-flash | xiaomi | MIT | 1,450.93 | 1,444.44 | 1,457.42 | 10.96 | 10,917 | overall | 2026-10-15 |
| 78 | gpt-5.3-chat-latest | openai | Proprietary | 1,450.04 | 1,445.81 | 1,454.28 | 4.67 | 34,345 | overall | 2026-10-15 |
| 79 | kimi-k2.5-thinking | moonshot | Modified MIT | 1,449.84 | 1,446.56 | 1,453.12 | 2.8 | 74,759 | overall | 2026-10-15 |
| 80 | claude-opus-4-1-20250805-thinking-16k | anthropic | Proprietary | 1,449.74 | 1,446.29 | 1,453.19 | 3.1 | 49,542 | overall | 2026-10-15 |
| 81 | ernie-5.0-preview-1203 | baidu | Proprietary | 1,448.54 | 1,442.04 | 1,455.04 | 11 | 9,806 | overall | 2026-10-15 |
| 82 | claude-opus-4-1-20250805 | anthropic | Proprietary | 1,448.08 | 1,445.11 | 1,451.05 | 2.3 | 77,238 | overall | 2026-10-15 |
| 83 | mimo-v2-pro | xiaomi | Proprietary | 1,447.79 | 1,443.13 | 1,452.45 | 5.66 | 25,447 | overall | 2026-10-15 |
| 84 | Step 5 Preview | stepfun | Proprietary | 1,447.27 | 1,439.76 | 1,454.78 | 14.68 | 7,116 | overall | 2026-10-15 |
| 85 | gpt-5.4-mini-high | openai | Proprietary | 1,447 | 1,443.25 | 1,450.75 | 3.66 | 63,449 | overall | 2026-10-15 |
| 86 | ernie-5.0-0110 | baidu | Proprietary | 1,446.19 | 1,442.3 | 1,450.09 | 3.95 | 36,487 | overall | 2026-10-15 |
| 87 | gemini-2.5-pro | Proprietary | 1,445.41 | 1,442.95 | 1,447.87 | 1.58 | 125,366 | overall | 2026-10-15 | |
| 88 | gpt-4.5-preview-2025-02-27 | openai | Proprietary | 1,444.63 | 1,438.92 | 1,450.34 | 8.49 | 14,547 | overall | 2026-10-15 |
| 89 | grok-4.7-xhigh | xai | Proprietary | 1,443.98 | 1,437.56 | 1,450.4 | 10.73 | 11,591 | overall | 2026-10-15 |
| 90 | qwen3.6-plus | alibaba | Proprietary | 1,443.06 | 1,438.92 | 1,447.19 | 4.45 | 48,165 | overall | 2026-10-15 |
| 91 | gpt-6-luna-max | openai | Proprietary | 1,442.81 | 1,438.19 | 1,447.44 | 5.57 | 49,617 | overall | 2026-10-15 |
| 92 | chatgpt-4o-latest-20250326 | openai | Proprietary | 1,442.6 | 1,439.79 | 1,445.4 | 2.05 | 81,914 | overall | 2026-10-15 |
| 93 | grok-4.3 | xai | Proprietary | 1,441.72 | 1,438.09 | 1,445.34 | 3.42 | 71,347 | overall | 2026-10-15 |
| 94 | qwen3.5-397b-a17b | alibaba | Apache 2.0 | 1,441.57 | 1,438.48 | 1,444.66 | 2.49 | 93,567 | overall | 2026-10-15 |
| 95 | glm-4.7 | zai | MIT | 1,441.33 | 1,435.27 | 1,447.38 | 9.55 | 12,253 | overall | 2026-10-15 |
| 96 | minimax-m3 | minimax | MiniMax Community License | 1,441.32 | 1,437.48 | 1,445.16 | 3.84 | 63,973 | overall | 2026-10-15 |
| 97 | inkling | thinky | Apache 2.0 | 1,440.87 | 1,436.56 | 1,445.18 | 4.84 | 40,648 | overall | 2026-10-15 |
| 98 | deepseek-v4-flash-high-preview | deepseek | MIT | 1,438.67 | 1,434.57 | 1,442.78 | 4.39 | 51,809 | overall | 2026-10-15 |
| 99 | gpt-5.1 | openai | Proprietary | 1,438.61 | 1,435.02 | 1,442.2 | 3.36 | 44,434 | overall | 2026-10-15 |
| 100 | qwen3.8-27b | alibaba | Apache 2.0 | 1,437.75 | 1,432.64 | 1,442.86 | 6.79 | 23,294 | overall | 2026-10-15 |