AJ-Bench (LLM-as-a-Judge) - Postgres: leaderboard
Verifying agent trajectories on AJ-Bench (2026), the Postgres subdomain (100 trajectories of MCPMark database tasks): the judge decides whether each trajectory succeeded and is scored against binary human-verified labels. In this setting the judge reads the trajectory only, with no tools or environment access. F1 in percent; higher is better.
Metric: F1 (%) on Postgres data-system tasks (MCPMark; 18 tasks, 100 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 February 2027. 13 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 73.2 | #119 | |
| 2 | Claude Opus 4.5 | 69.66 | #160 | |
| 3 | Gemini 2.5 Pro | 68.96 | #276 | |
| 4 | Claude Sonnet 4.5 | 68.36 | #261 | |
| 5 | DeepSeek V3.2 | 66.31 | #419 | |
| 6 | Kimi K2 0905 | 65.85 | #570 | |
| 7 | LongCat-Flash-Chat | 65.54 | #818 | |
| 8 | GPT-5 Mini | Low | 65.52 | #364 |
| 9 | GLM-4.6 | 64.94 | #498 | |
| 10 | Qwen 3 235B A22B | 64.32 | #628 | |
| 11 | GPT-5 | 62.84 | #178 | |
| 12 | Grok 4 | 59.57 | #319 | |
| 13 | GPT-5.1 | 57.53 | #240 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=aj-bench-llm-as-a-judge-postgres · How It Works · Data refreshed daily, snapshot 2026-11-21.