German Statutory QA - Multi-Provision Pre-Amendment (Vanilla): leaderboard
Metric: Correct outcomes (%; share of 84 expert-validated German statutory questions combining several provisions, at least one of which applies in an earlier version whose legal conclusion a Gemini 3 Flash Preview judge scores fully correct, model without retrieval or tools). Source: arxiv.org. Saturation forecast: Around March 2028. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 42.86 |
| 2 | DeepSeek V3.2 | 27.38 |
| 3 | Claude Opus 4.5 | 26.19 |
Interactive version: theaggregate.ai/benchmark?slug=german-statutory-qa-multi-provision-pre-amendment-vanilla · How It Works · Data refreshed daily, snapshot 2026-09-26.