X-PCR - Severity Grading: leaderboard

X-PCR (2026) tests multimodal models on ophthalmic diagnosis as a six-stage reasoning chain over six imaging modalities, with multiple-choice questions built from 51 public datasets and checked by ophthalmologists. Accuracy at the severity grading stage, in percent; higher is better.

Metric: Accuracy (%) at stage 5, grading disease severity, 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 June 2027. 21 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-561.65#185
2Gemini 2.5 Pro58.07#280
3InternVL3-32B54.82
4InternVL3-8B53.91#1795
5GPT-5 Nano53.89#912
6Qwen 2.5 VL 72B53.52#782
7Qwen 3 VL 30B A3B51.51#791
8GLM-4.5V50.47#756
9Lingshu-7B49.64#1646
10Gemini 2.5 Flash49.27#466
11Qwen 3 VL 8B49.09#927
12Qwen 2.5 VL 7B48.14#1878
13Claude Haiku 4.547.73#587
14HuaTuoGPT-Vision-34B47.59#1840
15MedGemma-27B-IT46.94#1359

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=x-pcr-severity-grading · How It Works · Data refreshed daily, snapshot 2026-11-22.