PSEBench - Evidence Citation: leaderboard

Metric: Evidence citation micro F1 (%, M3) over canonical clause and guidance identifiers, on correctly triaged cases with a gold legal basis, on the 5,074-case MN29 benchmark (3,455 complete, 1,362 missing-information and 257 uncertain cases built from clause cards of the Minnesota 29 reportable adverse health events), agentic environment where the model may ASK a GPT-5.2 information provider within 10 turns and then answers with a verdict, clause, legal evidence and rationale; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 15 models tracked.

Top models

#ModelScore
1GPT-5.574.3
2Claude Opus 4.767.4
3GPT-566.4
4Claude Sonnet 4.665.8
5Gemini 3.1 Pro (Preview)65.5
6Gemini 2.5 Flash64.9
7Qwen 3 235B A22B 2507 Instruct60.6
8DeepSeek R156.3
9Mistral Small 3.249.4
10GPT-OSS-120B45.4
11GPT-5 Nano41.1
12Llama 3.1 8B Instruct30.7

Interactive version: theaggregate.ai/benchmark?slug=psebench-evidence-citation · How It Works · Data refreshed daily, snapshot 2026-09-29.