Anthropic has released a new chemistry report highlighting that Claude Opus 4.7 can now compete with dedicated NMR software in inferring molecular structures from spectral data. This advancement means that Opus 4.7 not only assists chemists in interpreting data but can also reverse-engineer molecule structures from NMR spectra, a task typically performed solely by human experts. Importantly, this general-purpose model has made significant strides in predictive accuracy, achieving minimal hydrogen prediction errors and nearly matching the performance of specialized tools like MestReNova on carbon predictions. This development underscores how general-purpose AI models, without specific fine-tuning, are increasingly capable of handling specialized scientific tasks such as translating between molecular structures and their corresponding NMR spectra.

Anthropic: Anthropic is an AI research company that develops advanced large language models with a focus on safety and capability. In this news, the company shared findings from its science blog demonstrating that its latest model can handle specialized chemistry tasks involving NMR spectroscopy. The report positions the general-purpose system as competitive with dedicated tools in molecular analysis without any chemistry-specific training.
Claude Opus 4.7: Claude Opus 4.7 is Anthropic’s frontier general-purpose AI model optimized for complex reasoning across diverse domains. The news shows its strong performance on NMR spectroscopy, where it predicts spectra accurately and infers molecular structures from data, matching or exceeding specialist chemistry software on key tasks. This capability highlights its potential to automate aspects of chemical structure determination that previously required human chemists or purpose-built tools.

`json
{
“Scientific AI”: “General-purpose AI models are now able to perform complex scientific tasks like spectral analysis without the need for domain-specific fine-tuning.”,
“Model Evaluation”: “Current evaluations show that advanced AI models can perform similarly to established NMR software tools in critical prediction and inference tasks.”,
“Chemistry Research”: “AI technology is now capable of automating the process of converting NMR data into molecular structures, a task that typically requires expert chemist input or specialized software.”
}
`