ClawBench v2: leaderboard
ClawBench v2 (TIGER-Lab) — expanded second-generation Claw agentic environment suite; reward rate across scripted episodes.
Metric: Reward Rate (%). Source: huggingface.co. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 44.6 | #49 |
| 2 | GPT-5.5 | 35.4 | #31 |
| 3 | GLM-5.1 | 34.6 | #97 |
| 4 | DeepSeek V4 Pro | 33.9 | #102 |
| 5 | DeepSeek V4 Flash | 2.3 | #137 |
| 6 | GLM 4.5 Air | 2.3 | #392 |
| 7 | MiniMax-M2.5 | 1.5 | #303 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=clawbench-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.