VIPER (Veterinary Pathology) - Free-Text: leaderboard
Free-text questions of VIPER (Harvard Medical School and Mass General Brigham, 2026): open answers about rat histology images graded against the pathologist answer from 0 to 100 by a GPT-5.4 judge with a per-question rubric; higher is better.
Metric: Free-text score (0-100), a composite rubric grade by a GPT-5.4 judge against the pathologist answer, over the free-text items of 1,251 questions written and validated by board-certified veterinary pathologists on 419 H&E-stained rat histology images from seven organ systems. Source: arxiv.org. Saturation forecast: Around February 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 55.1 | #123 |
| 2 | Qwen 3.5 27B | 50.6 | #359 |
| 3 | GPT-5.4 Mini | 50 | #389 |
| 4 | Gemma 4 31B (IT) | 48.3 | #342 |
| 5 | Claude Sonnet 4.6 | 42.8 | #161 |
| 6 | GPT-5.4 Nano | 27.7 | #635 |
| 7 | MedGemma 1.5 4B | 26 | #2562 |
| 8 | Gemini 2.5 Flash | 25.2 | #453 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=viper-veterinary-pathology-free-text · How It Works · Data refreshed daily, snapshot 2026-11-13.