EuroLLM

weights

open(Apache-2.0, BF16 safetensors, ungated)

data

documented-not-released(EuroWeb-2512 web subset released, remaining parallel/curated/synthetic and long-context components documented in the report but withheld)

code

closed(only deep-spin/eurollm-eval evaluation harness is EuroLLM-specific, deep-spin/Megatron-LM-pretrain a generic year-stale Megatron-LM fork with no EuroLLM configs)

checkpoints

open(intermediate Megatron-Core checkpoints per pretraining phase, eurollm-22b/9b/1.7b-mcore-phase1/2/3)

license

Apache-2.0(OSI)

EuroLLM ships ungated Apache-2.0 BF16 weights and publishes intermediate Megatron-Core checkpoints for every pretraining phase, matching what OLMo and Apertus release. Only the EuroWeb-2512 web subset of its training data is public -- the parallel, code, math, synthetic-math, book and long-context components are documented but withheld -- and the only EuroLLM-specific code released is an evaluation harness, not the training pipeline. That mix of open weights, partly documented data and unreleased training code keeps it at open weights rather than open source.

• https://arxiv.org/pdf/2602.05879 recorded 2026-10-14

  Technical report lists the pretraining corpus components (web, parallel, code, math, synthetic math, Wikipedia, ArXiv, Books, Apollo, Cosmopedia, ~60B long-context) and cites github.com/deep-spin/Megatron-LM-pretrain (pre-training) and github.com/deep-spin/eurollm-eval (evaluation) as codebases. Only the web subset (EuroWeb-2512) and the eval harness are actually released; the pretraining repo is a generic Megatron-LM fork (independently verified: last commit 2025-01-06, no EuroLLM configs).

• https://huggingface.co/api/datasets/utter-project/EuroWeb-2512 recorded 2026-10-14

  `private: false`, `gated: false`, size category 1B<n<10B, no license tag; the released multilingual web pretraining subset for the 2512 release. Establishes that a documented subset of the corpus is released while the remainder is not.

• https://huggingface.co/api/models/utter-project/EuroLLM-22B-2512 recorded 2026-10-14

  `private: false`, `gated: false`, `license: apache-2.0`, `safetensors.total` 22,637,328,384 BF16 parameters, LlamaForCausalLM. Establishes that the weights are downloadable, ungated and OSI-licensed.

• https://huggingface.co/api/models/utter-project/eurollm-22b-mcore-phase1 recorded 2026-10-14

  `private: false`, `gated: false`; a public Megatron-Core intermediate checkpoint for the 22B model's first pretraining phase, one of the eurollm-22b/9b/1.7b-mcore-phase1/2/3 series. Establishes checkpoints:open.

• https://huggingface.co/utter-project/EuroLLM-22B-2512 recorded 2026-10-14

  Model card "Model Card for EuroLLM-22B": "License: Apache License 2.0"; 22B parameter multilingual transformer trained from scratch on ~4T tokens; describes the three pretraining phases (initial 3.6T, annealing 400B, annealing-to-zero 100B) and the data sources (web, parallel, Wikipedia, Arxiv, books, math, code, Apollo) without releasing the non-web components.

Adoption is measured as trailing-30-day Hugging Face downloads summed across the ten declared EuroLLM SKUs -- base and instruct variants at 22B, 9B and 1.7B, current and preview -- on the utter-project org. Because downloads move from month to month across that many separate checkpoints, the reading carries only moderate confidence.

• https://huggingface.co/api/models?author=utter-project&full=false&limit=100 recorded 2026-09-04

  Trailing-30-day `downloads` per EuroLLM SKU on the utter-project org, summing to 127,672 across the ten declared base and instruct variants (22B/9B/1.7B, current and preview).

EuroLLM-22B beats Apertus, the strongest fully open European peer, on eight of ten English benchmarks in the technical report's own comparison, and its multilingual coverage across 35 languages is unusually strong. It still trails frontier open-weights models on reasoning and knowledge overall, so the win over Apertus keeps it even with that peer rather than a step ahead.

• https://arxiv.org/pdf/2602.05879 recorded 2026-09-04

  Table 3: EuroLLM-22B-Instruct English scores (IFEval 67.2, Hellaswag 69.7, MMLU 69.8, MMLU-Pro 50.8, BBH 55.3, ARC-C 89.8, GPQA 26.8, GSM8K 85.5, MATH-500 54.5, HumanEval 53.9) against Apertus, OLMo-3, Llama-3, Gemma-3 and Qwen-3 baselines; prose "EuroLLM-22B is the strongest model among the fully open European systems" and "still trails the very best open-weights models overall".