ClawBench (OpenClaw) - Social: leaderboard
ClawBench (2026, NAIL Group) tests web agents on everyday write-heavy online tasks such as purchases, bookings and applications on live websites, stopping the final state-changing request and judging each run against a human reference trajectory. This board is the task success rate in percent on the social tasks with agents in the OpenClaw harness; higher is better.
Metric: Success rate (%) on the 18 social tasks of the 153-task ClawBench paper evaluation (agents in the OpenClaw browser harness on live production websites, the final state-changing request intercepted before it reaches the server, each run judged pass or fail by an Agent-as-Judge (Claude Sonnet 4.6) against a human reference trajectory); higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 8 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | High | 38.9 | #173 |
| 2 | Qwen 3.5 (ClawBench checkpoint unspecified) | 22.2 | ||
| 3 | Claude Haiku 4.5 | 16.7 | #583 | |
| 4 | GLM-5 | 16.7 | #269 | |
| 5 | Gemini 3 Flash | 11.1 | #175 | |
| 6 | Gemini 3.1 Pro (Preview) | 5.6 | #99 | |
| 7 | Gemini 3.1 Flash Lite | 0 | #371 | |
| 8 | GPT-5.4 | High | 0 | #135 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=clawbench-openclaw-social · How It Works · Data refreshed daily, snapshot 2026-11-21.