OpenClaw Arena Model Leaderboard: leaderboard
Personal AI agent benchmark evaluating frontier models across real-world OpenClaw-style tasks.
Metric: Avg Score (imported). Source: openclaw-arena.github.io. Source: benchmarklist.com. Saturation forecast: Around December 2026. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.5 | 67.4 | #83 |
| 2 | GPT-5.2 | 64.4 | #110 |
| 3 | Gemini 2.5 Pro | 63.8 | #152 |
| 4 | GPT-5.1 | 62.5 | #135 |
| 5 | Claude Sonnet 4.6 | 62.4 | #90 |
| 6 | Claude Opus 4.6 | 61.7 | #63 |
| 7 | Gemini 3.1 Pro (Preview) | 61 | #58 |
| 8 | GPT-5 Mini | 60.8 | #186 |
| 9 | GPT-4.1 | 60.5 | #248 |
| 10 | Claude Sonnet 4 | 57.8 | #199 |
| 11 | Claude Sonnet 4.5 | 57.5 | #142 |
| 12 | GPT-4o | 52.2 | #343 |
| 13 | GPT-4o Mini | 42.1 | #599 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openclaw-arena-model-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.