BioTool - Ensembl: 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 461 Ensembl queries of the test split). Source: arxiv.org. Saturation forecast: Around April 2027. 12 models tracked.

Top models

#ModelScore
1Gemini 3 Pro77.3
2Claude Sonnet 4.577
3GPT-5.1 Codex74.5
4GPT-5.172.8
5Qwen 2.5 7B Instruct65.3
6Qwen 3 4B 2507 Instruct64
7Qwen 3 8B63.5
8Llama 3.1 8B Instruct58.8

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