PetQA-Bench - Text Factuality: leaderboard
Veterinary consultation questions from real pet owners, answered text-only: claim-level factuality against veterinary references. Higher is better. Source: arxiv.org/abs/2609.04598.
Metric: Factuality (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 | 56.9 | #259 |
| 2 | GPT-4.1 Mini | 55.1 | #364 |
| 3 | Gemini 2.5 Pro | 55 | #159 |
| 4 | Gemini 2.5 Flash | 54.1 | #254 |
| 5 | Qwen 3 32B | 53 | #449 |
| 6 | Gemma 3 27B | 51.9 | #606 |
| 7 | Gemma 2 9B | 51.8 | #924 |
| 8 | Qwen 3 8B | 51.4 | #679 |
| 9 | Gemma 3 12B | 51 | #674 |
| 10 | A.X-4.0-Light | 50.2 | #777 |
| 11 | Gemma 2 27B | 48.9 | #805 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=petqa-bench-text-factuality · How It Works · Data refreshed daily, snapshot 2026-11-06.