Intology's Locus automates molecular design, outperforming Claude Code and Codex

AIIntology says its autonomous research system Locus sets the frontier for automating molecular design, including property prediction, Boltz-2 potency prediction, and antibody design workflows. On held-out test data, Locus held up better than Claude Code and Codex. The system adapts models like Boltz-2 to private targets with limited data, work that usually requires expensive ML expertise.

AIRohan Paul reports that Intology's automated research system Locus outperformed state-of-the-art baselines and LLM agents on molecular design tasks. The tasks include property prediction, fine-tuning Boltz-2 for small-molecule potency, and building antibody design workflows. Locus is described as running experiments in parallel on GPUs and keeping what works, and the post includes a benchmark figure comparing it with Codex, Claude Code, Boltz-2 API, BoltzGen, and ESMFold2.