TIGER-AI-Lab / ClawBench
Open-source benchmark for browser AI agents on daily tasks.
Why this job: says: benchmark; topics: evaluation, llm-evaluation, benchmark
What it does
ClawBench is an open-source benchmark, published in Findings of EMNLP 2026, that evaluates AI browser agents on everyday online tasks — booking travel, ordering food, applying for jobs, managing email — across live websites. It records five layers of evidence and supports request-based scoring and trajectory evaluation against human references. In the historical V1 paper evaluation, the strongest evaluated agent completed about 1 in 3 tasks.
Get started
uv tool install clawbench-eval
clawbench- Maintenance
- Active
- Last commit 4 days ago
- Stars
- 1,240
- +51 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.11.0
- Released 35 days ago
- Built with
- Python
- 86% of the code
Is it looked after?
- Contributors
- 10
- Last 90 days
- Issues
- 51% closed
- 109 still open
- Pull requests
- 200 merged
- 27 waiting
- Releases
- 19
- 19 in the last year
- Contributing guide (opens in new tab)
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| HowieHwong / TrustLLM | 635 | 36 days ago | MIT | No downloads |
| AMAP-ML / LoopArena | 154 | 26 days ago | Apache-2.0 | No downloads |
| StartLuxLabs / DecisionPivot | 154 | 19 days ago | Apache-2.0 | No downloads |
More: languages, topics and project record
Languages
- Python 86%
- Shell 13%
- JavaScript 1%
- Dockerfile <1%
Topics
- agent-evaluation
- agentic-ai
- ai-agent-benchmark
- ai-agents
- benchmark
- browser-agent
- browser-automation
- browser-use
- chrome-agent
- chrome-extension
- computer-use
- dataset
Project record
- JobEvals and observability
- StatusFeatured
- Commits555
- Forks75
- Watching14
- Stars in 28 days+251
- CreatedApr 10, 2026
- First seen by usNov 7, 2026
- CheckedNov 7, 2026