AJ-Bench (LLM-as-a-Judge) - FileSystem: leaderboard
Metric: F1 (%) on FileSystem data-system tasks (MCPMark; 24 tasks, 129 trajectories), averaged over three runs, with the judge reading the trajectory only (no tools or environment access), of AJ-Bench, judging whether agent trajectories succeeded against binary human-verified labels (155 tasks, 516 trajectories), default model configurations; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4 | 75.7 |
| 2 | Gemini 3 Pro (Preview) | 75.69 |
| 3 | Claude Sonnet 4.5 | 69.26 |
| 4 | Gemini 2.5 Pro | 66.1 |
| 5 | Claude Opus 4.5 | 66.06 |
| 6 | Qwen 3 235B A22B | 64.66 |
| 7 | GLM-4.6 | 60.86 |
| 8 | GPT-5 Mini (Low) | 60.41 |
| 9 | DeepSeek V3.2 | 60.31 |
| 10 | LongCat-Flash-Chat | 59.13 |
| 11 | GPT-5 | 59.09 |
| 12 | Kimi K2 0905 | 55.96 |
| 13 | GPT-5.1 | 46.27 |
Interactive version: theaggregate.ai/benchmark?slug=aj-bench-llm-as-a-judge-filesystem · How It Works · Data refreshed daily, snapshot 2026-10-07.