AI model rankings
Understand capability differences before choosing AI resources.
LiveBench 2.0 · Multidimensional capabilities
Public evaluation snapshots · Ranked separately by dimension
LiveBench 2.0 · 推理 · 发布 20261025 · Scores range from 0–100; higher is better. Missing scores are excluded, and ties share a rank. Search preserves original ranks.
| Rank | Model and version | Score | Evidence type |
|---|---|---|---|
| 1 | gemini-4-argon官方榜单收录 · 评测设置见来源Model details ↗ | 95.1 | Leaderboard listing |
| 2 | gpt-6-astra官方榜单收录 · 评测设置见来源Model details ↗ | 92.7 | Leaderboard listing |
| 3 | gpt-6.1-sol官方榜单收录 · 评测设置见来源Model details ↗ | 92.6 | Leaderboard listing |
| 4 | claude-opus-5.5官方榜单收录 · 评测设置见来源Model details ↗ | 92.2 | Leaderboard listing |
| 5 | claude-fable-5.1官方榜单收录 · 评测设置见来源Model details ↗ | 91.7 | Leaderboard listing |
| 6 | claude-sonnet-5.5官方榜单收录 · 评测设置见来源Model details ↗ | 91.6 | Leaderboard listing |
| 7 | grok-4.6官方榜单收录 · 评测设置见来源Model details ↗ | 91.2 | Leaderboard listing |
| 8 | kimi-k3.1-pro官方榜单收录 · 评测设置见来源Model details ↗ | 91.1 | Leaderboard listing |
| 9 | kimi-k3官方榜单收录 · 评测设置见来源Model details ↗ | 90.7 | Leaderboard listing |
| 10 | gpt-5.6-terra官方榜单收录 · 评测设置见来源Model details ↗ | 90.6 | Leaderboard listing |
| 10 | muse-wonder-1.0官方榜单收录 · 评测设置见来源Model details ↗ | 90.6 | Leaderboard listing |
| 12 | muse-spark-1.3官方榜单收录 · 评测设置见来源Model details ↗ | 89.7 | Leaderboard listing |
| 13 | gemini-3.8-flash官方榜单收录 · 评测设置见来源Model details ↗ | 89.3 | Leaderboard listing |
| 14 | deepseek-v4.1-flash-1017官方榜单收录 · 评测设置见来源Model details ↗ | 89.1 | Leaderboard listing |
| 15 | gpt-6.1-terra官方榜单收录 · 评测设置见来源Model details ↗ | 88.7 | Leaderboard listing |
| 16 | qwen3.8-max官方榜单收录 · 评测设置见来源Model details ↗ | 88.2 | Leaderboard listing |
| 17 | mistral-large-4官方榜单收录 · 评测设置见来源Model details ↗ | 83.9 | Leaderboard listing |
| 18 | step-5-preview官方榜单收录 · 评测设置见来源Model details ↗ | 83.8 | Leaderboard listing |
| 19 | minimax-m3.1-flash官方榜单收录 · 评测设置见来源Model details ↗ | 82.9 | Leaderboard listing |
| 20 | grok-4.7官方榜单收录 · 评测设置见来源Model details ↗ | 82.7 | Leaderboard listing |
| 21 | claude-haiku-5.5官方榜单收录 · 评测设置见来源Model details ↗ | 82.4 | Leaderboard listing |
| 22 | gpt-6-luna官方榜单收录 · 评测设置见来源Model details ↗ | 81.8 | Leaderboard listing |
| 23 | qwen3.8-27b官方榜单收录 · 评测设置见来源Model details ↗ | 80.0 | Leaderboard listing |
| 24 | mimo-v2.6-pro官方榜单收录 · 评测设置见来源Model details ↗ | 79.6 | Leaderboard listing |
| 25 | inkling官方榜单收录 · 评测设置见来源Model details ↗ | 78.3 | Leaderboard listing |
Source evidence and evaluation methodology
Source:LiveBench 2.0 · Multidimensional capabilities · Public numerical facts · No data relicensing declared. We organize original scores without creating cross-benchmark composite scores. Listing a result does not mean we independently reproduced it.
Attribution: LiveBench. We record public model scores and source links only, without copying test questions, site content, or code. We do not apply an old repository license to the new leaderboard.
Source snapshot:1115296907a94ada817509ee2c0f3748 · Source release version：20261025 · Original content SHA-256:1b8e3148febc91c92cdbc86c77241922295d6d29e74354dbfe4e32a5e57f2bfe
Fetched on demand and published after admin review, not in real time. Undisclosed evaluation versions are not inferred.Different dimensions and test settings cannot be compared directly.