BLOOM

BLOOM is a 176 billion parameter language model built by BigScience, a year-long collaboration of over a thousand researchers coordinated through Hugging Face with compute donated by the French government’s Jean Zay supercomputer. It launched in July 2022 as an answer to closed models like GPT-3, trained on the ROOTS corpus covering 46 natural languages and 13 programming languages, with a deliberate emphasis on underrepresented languages that most large labs ignored at the time. The project published its training data, code, and a full technical report alongside the weights, which was rare for a model at that scale.

BLOOM is released under the RAIL License, a permissive-but-restricted license that allows commercial use while banning specific harmful applications, an approach BigScience helped popularize. Its 2,048-token context window and dense architecture look dated next to current frontier models, and it has largely been superseded by more efficient open-weight families such as Llama and Mistral, but it remains a reference point in discussions of multilingual coverage and open, community-governed model development.