Open-model cases
OLMo
Ai2 releases the OLMo family with its pretraining data, training code, intermediate checkpoints and logs alongside the weights. The first OLMo appeared in February 2024, OLMo 2 that November, and Olmo 3 in November 2025 with 7B and 32B Base, Instruct and Think variants under Apache 2.0. It is the case where all four things worth checking separately — code, weights, training disclosure and license — are actually published, which makes it a reference point against families that release weights alone.
Key terms
- fully open
- open training data
- intermediate checkpoints
- training logs
Openness profile
Code and weights are under Apache 2.0. The Dolma pretraining corpus is under ODC-BY and remains subject to the licenses of its original sources, so open data here does not mean unrestricted data. Training code, intermediate checkpoints and logs are published, so the training disclosure can be inspected rather than inferred.
"Open source", "open model", and "open weights" are not interchangeable. Review code, weights, training information, and license separately for each release.
Connected across the project
Related papers
Historical context
Related sections
Primary sources
This is a curated technical map, not a claim of comprehensive coverage.