WeClawArena - Bidding: leaderboard
Task success rate on the Bidding domain of WeClawArena (2026), computed over all five variants of each of the 12 base tasks (one benign control and four attack-vector variants). Percent, higher is better.
Metric: Task success rate (%; deterministic check of the final multi-workspace state over all five variants of each of the 12 Bidding base tasks, one no-attacker control and four attack-vector variants; each model drives every owner agent in the Dockerized OpenClaw runtime; rows with missing task-success fields count as failures). Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 55 | #75 |
| 2 | Claude Sonnet 4.5 | 51.7 | #244 |
| 3 | Kimi K2.5 | 40 | #252 |
| 4 | Claude Opus 4.1 | 6.7 | #268 |
| 5 | DeepSeek V3.2 | 5 | #403 |
| 6 | Qwen 3 235B A22B | 3.3 | #622 |
| 7 | Kimi K2 (Thinking) | 3.3 | #435 (Kimi K2) |
| 8 | Qwen 3 32B | 1.7 | #960 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=weclawarena-bidding · How It Works · Data refreshed daily, snapshot 2026-11-12.