C3LM

C3LM is a chemical plausibility-aware language model trained on 45.6M verified reactions, achieving state-of-the-art single-step retrosynthesis via Top-K prompting. It was developed to enhance synthetic route prediction accuracy.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.