Skip to content

Anthropic

Claude advances chemical spectrum analysis

Written by an AI model from the linked sources and checked against them by a second model. How we work

Image: anthropic.com

Anthropic tested Opus 4.7 on NMR spectroscopy tasks. The model predicted hydrogen and carbon peak positions for 20 compounds, competing with ChemDraw and MestReNova.

In direct spectrum prediction from structure, Opus 4.7 matched industry tools in accuracy. The 0.079 ppm average error for hydrogen fell within acceptable bounds. Claude also succeeded in inverse tasks: deducing chemical structure from spectral data.

Evaluations used 15 complex molecules. The model reconstructed simple compound structures without hints; for complex cases, providing raw materials guided it. Anthropic plans to expand AI for Science, focusing on reading chemical schemes and synthesis planning.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.