LiveBench Reasoning: leaderboard
Metric: Score. Source: livebench.ai. Saturation forecast: Around November 2027. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 95.1 | #7 |
| 2 | GPT-6 Astra | 92.7 | #5 |
| 3 | GPT-6.1 Sol | 92.6 | #9 |
| 4 | Claude Opus 5.5 | 92.2 | #3 |
| 5 | GPT-5.6 Sol | 91.7 | #18 |
| 6 | Claude Fable 5.1 | 91.7 | #8 |
| 7 | Claude Sonnet 5.5 | 91.6 | #11 |
| 8 | Claude Opus 5 | 91.2 | #12 |
| 9 | Kimi K3 | 90.7 | #29 |
| 10 | GPT-5.6 Terra | 90.6 | #42 |
| 11 | Grok 4.6 | 90.5 | #27 |
| 12 | Muse Spark 1.2 | 90 | #62 |
| 13 | GPT-5.5 | 89.7 | #30 |
| 14 | Claude Fable 5 | 89.7 | #15 |
| 15 | Muse Spark 1.3 | 89.7 | #25 |
No result here: #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-14.