FailBench - robometersim: leaderboard
Metric: Balanced Accuracy (%). Source: arxiv.org. Saturation forecast: Around December 2026. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 83 |
| 2 | Gemini 3 Flash | 82 |
| 3 | GPT-4o | 72 |
| 4 | Qwen 3 VL 8B (Thinking) | 71 |
| 5 | Qwen 3 VL 4B (Thinking) | 70 |
Interactive version: theaggregate.ai/benchmark?slug=failbench-robometersim · How It Works · Data refreshed daily, snapshot 2026-09-24.