Browser Use Bench v2.1: leaderboard
Metric: Mean weighted rubric score (0-100). Source: browser-use.com. Saturation forecast: Around May 2027. 17 models tracked. Updated 18 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-6.1 Sol | High | 81.3 | #12 |
| 2 | GPT-6 Sol | High | 79.4 | #30 |
| 3 | GPT-6.1 Terra | High | 79 | #22 |
| 4 | GPT-6 Astra | Medium | 77.9 | #7 |
| 5 | Kimi K3.1 Pro | High | 76.7 | #32 |
| 6 | DeepSeek V4.1 Flash (1017) | Max | 75 | #43 |
| 7 | Claude Opus 5.5 | High | 75 | #6 |
| 8 | Grok 4.7 | High | 73.3 | #48 |
| 9 | Claude Sonnet 5.5 | High | 72.7 | #21 |
| 10 | Claude Fable 5.1 | High | 72 | #17 |
| 11 | DeepSeek V4.1 Flash | Max | 70.9 | #67 |
| 12 | GPT-5.6 Luna | xHigh | 69.6 | #129 |
| 13 | Claude Haiku 5.5 | xHigh | 69.5 | #75 |
| 14 | GPT-6 Luna | Max | 69.2 | #102 |
| 15 | Claude Opus 5 | 66.8 | #24 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=browser-use-bench-v2-1 · How It Works · Data refreshed daily, snapshot 2026-11-18.