Claude Mythos 5.1 protein designs receive reported external wet-lab validation

What changed. Anthropic reports nearly 50% binder hit rate across 12 targets and tenfold affinity improvement on three targets versus prior competition winners.

Technique / discovery. Tool-using protein design and folding, external binding assays, GPU kernel generation, and caching.

Why it matters. Scientific-agent claims become stronger when they terminate in physical assays, but reproducibility still requires full artifacts and negative results.

Application. Protein engineering and computational biology

Evidence maturity. Technical report (source tier A)

Identified bottleneck. Sequences, protocols, selection funnel, negative results, matched design budgets, and kernel code were unavailable at review.

Caveat / evidence note. First-party model-release report with stated external validation; not peer reviewed or independently reproduced.

Review status. Reviewed. User requested: Yes.

• Anthropic (opens in a new tab)

Protein design validation protocol and computational smoke test — published with the 2026-09-02 briefing edition.