EuroLLM-9B
Europe's open multilingual LLM covering all 24 official EU languages
About
EuroLLM-9B, released December 9, 2024, is the flagship of a European research consortium effort to build open models for all 24 official EU languages. The 9B model trained on roughly 4 trillion multilingual tokens across 35 languages, and the release is unusually open: Apache 2.0 weights in base and Instruct variants, a full technical report, the EuroFilter data-filtering classifier, and the EuroBlocks-Synthetic post-training dataset. The team reported it the most capable open European-made LLM of its size at release. Limitations are honest ones: a 4,096-token context window, uneven quality across the 35 languages, and raw reasoning well below frontier models. A smaller 1.7B sibling and a later 22B followed.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Team-reported: led its size class on multilingual EU benchmarks at release (team-reported)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: Released 2024-12-09 per technical report (arXiv 2506.04079); Apache 2.0 confirmed; unusually open (weights, report, EuroFilter, EuroBlocks-Synthetic); 9B, ~4T multilingual tokens, 35 languages incl. all 24 EU official (team specs); 4K context and uneven per-language quality noted as limitations.