FrontierBioML benchmark: 25 long-horizon comp bio tasks, Claude Mythos 5.2 leads at 40.3

afurgs · x · 2026-11-21

FrontierBioML introduces 25 long-horizon ML tasks in computational biology, evaluating 11 models with runs up to 6 hours each. Agents submit a prediction script that organizers run on a held-out test set.

Leaderboard: Claude Mythos 5.2 leads at 40.3, followed by Gemini 4 Carbon (39.0) and Gemini 4 Argon (36.8).

Original post →

More from Models

Models channel →