CheXpercept — leaderboard
Metric: Stage 1 (End-to-End) (self-reported). Source: benchmarklist.com. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 27B | 92.2 |
| 2 | Qwen 3.5 122B A10B | 91.3 |
| 3 | Gemini 3.1 Pro (Preview) | 90 |
| 4 | Gemma 4 31B | 88.5 |
| 5 | Gemini 3.1 Flash | 87.1 |
| 6 | GPT-5.4 | 84.3 |
| 7 | GPT-5.4 Nano | 76.9 |
| 8 | GLM-4.6V | 69.3 |
Interactive version: theaggregate.ai/benchmark?slug=chexpercept · How the rankings work · Data refreshed daily, snapshot 2026-07-22.