LiveBench Coding: leaderboard
Metric: Score. Source: livebench.ai. Saturation forecast: Around November 2027. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 98.3 | #15 |
| 2 | Claude Fable 5.1 | 97.7 | #8 |
| 3 | Claude Opus 5.5 | 96.3 | #3 |
| 4 | GPT-6.1 Sol | 96 | #9 |
| 5 | Claude Sonnet 5.5 | 94.7 | #11 |
| 6 | Grok 4.5 | 94.3 | #48 |
| 7 | GPT-6 Astra | 93.7 | #5 |
| 8 | GPT-6 Sol | 93.1 | #16 |
| 9 | Claude Opus 5 | 90.8 | #12 |
| 10 | Muse Spark 1.3 | 90.2 | #25 |
| 11 | GPT-5.6 Sol | 89.7 | #18 |
| 12 | Gemini 4 Argon | 89.7 | #7 |
| 13 | Grok 4.6 | 89.1 | #27 |
| 14 | GPT-5.6 Terra | 86.2 | #42 |
| 15 | DeepSeek V4.1 Flash | 85.4 | #39 |
No result here: #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-coding · How It Works · Data refreshed daily, snapshot 2026-10-14.