最后更新
Dec 5, 2025, 11:39 PM
探索根据社区投票和性能指标排名的顶级AI模型
模型总数
24
数据库中可用
顶级模型
Anthropicclaude-opus-4-5-20251101-thinking-32k
排名第一
最高分
1511
最高评分
| # | 模型 | 评分 | 投票数 | 置信区间 | 组织 |
|---|---|---|---|---|---|
| 🥇 | Anthropicclaude-opus-4-5-20251101-thinking-32k | 1511.00 | 2,323 | ±14 | Anthropic |
| 🥈 | gemini-3-pro | 1476.00 | 7,154 | ±10 | |
| 🥉 | Anthropicclaude-opus-4-5-20251101 | 1472.00 | 2,377 | ±14 | Anthropic |
| 4 | gpt-5-medium | 1399.00 | 3,943 | ±12 | OpenAI |
| 5 | Anthropicclaude-sonnet-4-5-20250929-thinking-32k | 1398.00 | 6,217 | ±9 | Anthropic |
| 6 | gpt-5.1-medium | 1395.00 | 3,429 | ±11 | OpenAI |
| 7 | Anthropicclaude-opus-4-1-20250805 | 1392.00 | 6,028 | ±9 | Anthropic |
| 8 | Anthropicclaude-sonnet-4-5-20250929 | 1387.00 | 7,311 | ±9 | Anthropic |
| 9 | glm-4.6 | 1366.00 | 5,806 | ±10 | Z.ai |
| 10 | gpt-5.1 | 1354.00 | 5,270 | ±10 | OpenAI |
| 11 | MoonshotAIkimi-k2-thinking-turbo | 1350.00 | 5,118 | ±10 | Moonshot |
| 12 | gpt-5.1-codex | 1341.00 | 3,614 | ±11 | OpenAI |
| 13 | Minimaxminimax-m2 | 1316.00 | 5,783 | ±10 | MiniMax |
| 14 | deepseek-v3.2-exp | 1293.00 | 5,154 | ±10 | DeepSeek A |
| 15 | qwen3-coder-480b-a35b-instruct | 1289.00 | 5,972 | ±9 | Alibaba |
| 16 | Anthropicclaude-haiku-4-5-20251001 | 1285.00 | 5,992 | ±9 | Anthropic |
| 17 | KwaiKAT-Coder-Pro-V1 | 1264.00 | 1,943 | ±15 | KwaiKAT |
| 18 | gpt-5.1-codex-mini | 1252.00 | 1,564 | ±16 | OpenAI |
| 19 | grok-4-1-fast-reasoning | 1229.00 | 2,978 | ±13 | xAI |
| 20 | gemini-2.5-pro | 1213.00 | 3,504 | ±12 | |
| 21 | grok-4.1-thinking | 1205.00 | 1,258 | ±19 | xAI |
| 22 | grok-4-fast-reasoning | 1153.00 | 943 | ±22 | xAI |
| 23 | grok-code-fast-1 | 1143.00 | 1,014 | ±21 | xAI |
| 24 | devstral-medium-2507 | 1103.00 | 1,031 | ±21 | Mistral |
数据每小时更新 • 显示 24 个模型
数据来源：LM BASE 排行榜