PatRe - Office Action Quality (OA-RS): leaderboard
Metric: Overall office action quality on a 1-10 scale: mean of the soundness, clarity, constructiveness, completeness and legal-style ratings of a Gemini-3.1-Flash-Lite judge at temperature 0, office action generation under retrieval simulated: the model receives a noisy pool of BM25-retrieved references mixed with ground-truth ones, over PatRe (480 recent USPTO patent examination histories across all eight IPC sections, with office actions, rebuttals, claim versions and cited references), temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around January 2028. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Mini | 5.39 |
| 2 | Gemini 2.5 Flash | 4.37 |
| 3 | Qwen 3.5 27B | 4.3 |
| 4 | Qwen 3.5 9B | 4.11 |
| 5 | Gemma 3 27B (IT) | 3.68 |
| 6 | GPT-4o Mini | 3.59 |
| 7 | Gemma 3 12B (IT) | 3.51 |
| 8 | Llama 3.3 70B Instruct | 3.5 |
| 9 | Llama 3.1 8B Instruct | 3.19 |
Interactive version: theaggregate.ai/benchmark?slug=patre-office-action-quality-oa-rs · How It Works · Data refreshed daily, snapshot 2026-10-07.