X-PCR - Resident-Level Questions: leaderboard
Metric: Accuracy (%) on the questions ophthalmologists rated at resident difficulty, on X-PCR's ophthalmic multiple-choice VQA (26,415 images from 51 public datasets, six imaging modalities, 52 diseases), zero-shot, temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 80.64 |
| 2 | GLM-4.5V | 77.91 |
| 3 | Gemini 2.5 Pro | 75.71 |
| 4 | Qwen 2.5 VL 7B | 74.81 |
| 5 | MedGemma-27B-IT | 74.19 |
| 6 | Gemini 2.5 Flash | 73.29 |
| 7 | GPT-5 Nano | 73.23 |
| 8 | InternVL3-8B | 70.73 |
| 9 | Claude Haiku 4.5 | 65.5 |
Interactive version: theaggregate.ai/benchmark?slug=x-pcr-resident-level-questions · How It Works · Data refreshed daily, snapshot 2026-10-07.