AstroLLM

A domain-specialized language model for astronomy and astrophysics.

Retrieval-grounded. Tool-integrated. Openly built.

AstroLLM is building an open intelligence layer for astronomy — connecting the world's astronomical literature, databases, and observations through specialized language models that can retrieve, reason, cite, and teach.

What's coming

• AstroLLM Core — Fine-tuned 4B/8B models for astronomy Q&A with citations
• Training datasets — Curated SFT data from NASA ADS, SIMBAD, and the Exoplanet Archive
• Evaluation benchmarks — Custom astronomy-specific eval suites

Built on

NASA ADS · SIMBAD · NASA Exoplanet Archive · NED · PDS · Gaia

Links

• 🌐 astrollm.org
• 💻 GitHub

---

Models, datasets, and evaluation benchmarks will be published here as they're developed.