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

#ModelSettingScoreOverall rank
1Gemini 3 Pro (Preview)73.2#119
2Claude Opus 4.569.66#160
3Gemini 2.5 Pro68.96#276
4Claude Sonnet 4.568.36#261
5DeepSeek V3.266.31#419
6Kimi K2 090565.85#570
7LongCat-Flash-Chat65.54#818
8GPT-5 MiniLow65.52#364
9GLM-4.664.94#498
10Qwen 3 235B A22B64.32#628
11GPT-562.84#178
12Grok 459.57#319
13GPT-5.157.53#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.