OpenLLaMALLaMA reproduction with pretrained 3B, 7B, and 13B language-model weights for PyTorch and JAX.Apache-2.0
LLaMA reproduction with pretrained 3B, 7B, and 13B language-model weights for PyTorch and JAX.
OpenLLaMA provides pretrained language models that reproduce the LLaMA architecture using different training datasets. The project distributes 3B, 7B, and 13B parameter model families with weights for PyTorch through Hugging Face Transformers and for JAX through EasyLM. Its documentation includes loading examples, training-data details, and evaluation comparisons for the released models.
OpenLLaMA official README · Reviewed 2026-10-13
- 01 / Feature noteLoad pretrained weights
Load PyTorch checkpoints through Hugging Face Transformers or use the JAX weights with EasyLM.
Feature source ↗ - 02 / Feature noteChoose model sizes
Select released model families with 3B, 7B or 13B parameters, including separately documented v1 and v2 training mixtures.
Feature source ↗
GitHub stars
7,518
Detected license
Apache-2.0
Repository created
2023-04-28
Repository push
2023-07-16
2023-04-28Repository push
2023-07-16
GitHub stars
Unknown
i
Cached history uses a different month window; chart withheld until aligned.
Latest attempt: ok · Checked 2026-10-15T19:40:39.600225+00:00 · Last successful fetch 2026-10-15T19:40:39.600225+00:00 · Provider source
Hacker News stories
Unknown
i
OpenSSF Scorecard
Unknown
i
No result returned.
Checked 2026-10-27T12:07:11Z · Latest attempt: not found · Provider source
GitHub release
Unknown
i
No result returned.
Checked 2026-11-04T04:26:56Z · Latest attempt: not found · Provider source
Unknown
