BioTool - UniProt: leaderboard

Metric: BioTool score (%; mean MedCPT embedding similarity between the observation returned by the model's API call and the reference observation, 0 for a failed call; the 471 UniProt queries of the test split). Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.

Top models

#ModelScore
1Gemini 3 Pro87.8
2Claude Sonnet 4.587.3
3GPT-5.1 Codex80.5
4Qwen 2.5 7B Instruct79
5GPT-5.178.7
6Llama 3.1 8B Instruct77.5
7Qwen 3 8B73.1
8Qwen 3 4B 2507 Instruct71.9

Interactive version: theaggregate.ai/benchmark?slug=biotool-uniprot · How It Works · Data refreshed daily, snapshot 2026-09-25.