Allen Institute for AI
OLMo
AI2's fully-open initiative: weights, data and training code all under permissive licences.
Visit homepage ↗History & context
The Allen Institute for AI's OLMo line is the standard-bearer for fully-open AI: weights, training data, training code, and intermediate checkpoints are all released under permissive licences. The thesis is that reproducibility matters as much as the final weights.
OLMo 2 (November 2024) shipped at 7B and 13B sizes under Apache 2.0. The smaller scale relative to commercial frontiers reflects AI2's research-first priorities — these models are designed to be studied, modified and reproduced rather than to top leaderboards.
Flagship model
2 models in this family
Larger OLMo 2 release. Same fully-open philosophy as the 7B variant. The 13B size makes it more competitive with mainstream production-grade chat models.
- Context
- 4K
- License
- apache-2-0
- VRAM Q4
- 7.8 GB
Fully-open 7B model: weights, training data and code all released under permissive licences. Useful as a reference for reproducibility research and for teams that need full transparency on training data provenance.
- Context
- 4K
- License
- apache-2-0
- VRAM Q4
- 4.2 GB