RxEval — leaderboard

Metric: F1 (self-reported). Source: benchmarklist.com. 17 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)77.1
2GPT-5.471.9
3GPT-571.8
4Qwen 3.6 27B71.7
5GPT-5 Mini71.3
6Gemini 3 Flash (Preview)70.6
7Gemini 2.5 Pro70.2
8Gemini 2.5 Flash67.3
9GPT-4o58.9
10Qwen 2.5 72B Instruct56.1
11Gemma 3 27B55
12Qwen 2.5 7B Instruct53
13Llama 3.1 8B Instruct45.18

Interactive version: theaggregate.ai/benchmark?slug=rxeval · How the rankings work · Data refreshed daily, snapshot 2026-07-22.