Millions of main-sequence binary stars from Gaia BP/RP spectra

¹ Max-Planck-Institut für Astronomie, Königstuhl 17, 69117 Heidelberg, Germany
² Department of Astronomy, The Ohio State University, Columbus, OH 43210, USA
³ Center for Cosmology and AstroParticle Physics (CCAPP), The Ohio State University, Columbus, OH 43210, USA
⁴ Department of Astronomy, California Institute of Technology, 1200 E. California Blvd., Pasadena, CA 91125, USA
⁵ Key Lab of Space Astronomy and Technology, National Astronomical Observatories, Beijing 100101, PR China
⁶ Institute for Frontiers in Astronomy and Astrophysics of Beijing Normal University, Beijing 100875, PR China
⁷ University of Chinese Academy of Sciences, Beijing 100049, China
⁸ Zhejiang Lab, Hangzhou 311121, PR China

^★ Corresponding author: jdli@mpia.de

Received: 11 July 2025
Accepted: 11 October 2025

Abstract

We present an extensive catalog of likely main-sequence main-sequence (MSMS) binary stars systems derived from Gaia Data Release 3 BP/RP (XP) spectra through the comparison of single- and binary-star model fits. Leveraging the large sample of low-resolution Gaia XP spectra, we used a neural network to build a forward model for spectral luminosities of single stars as a function of stellar mass and photometric metallicity. Applying this model to XP spectra, we found that this model enables the identification of binaries with mass ratios between 0.5 and 1.0 and flux ratios larger than 0.1 as “poor” fits to the data, either in spectral shape or flux normalization. From an initial sample of 35 million stars within 1 kpc, we identified 14 million possible binary candidates, and a high-confidence “golden sample” of 1 million binary systems. This large homogeneous sample of spectrophotometry-based binaries enables population studies of luminous MSMS binaries and – in conjunction with kinematic or astrometric probes – permits identification of binaries with dark or dim companions, such as white dwarfs, neutron star and black hole candidates, improving our understanding of compact object populations.