Marin
Stanford CRFMStanford CRFM's Marin, a fully-open 'open lab' foundation model (Marin 8B Base/Instruct and 32B Base), built in JAX/Levanter with bit-for-bit reproducibility. Code, data, experiments, hyperparameters and training logs are all documented and released.
~12.7T-token training run with full data/code/logs and bit-reproducible training. Fully open but below OLMo 3 on fully-open evals; modest adoption. The distinguishing move is procedural rather than about the artifacts - every experiment is tracked as a public GitHub issue, so the record of how the model was reached is open as well as the recipe. Verified 2026-08-13 via the Marin announcement post, the marin-8b-base HF model card and the marin-community/marin repository.
Openness
5 high confidence- weights
- open(Apache-2.0)
- data
- open(documented mixtures)
- code
- open(marin + Levanter, JAX)
- checkpoints
- open
- reproducibility
- bit-for-bit
- license
- Apache-2.0(OSI)
Stanford CRFM 'open lab': Apache-2.0 weights plus documented code, data, experiments, hyperparameters and training logs, with bit-for-bit reproducible training.
- http://marin.community/blog/2025/05/19/announcement/ recorded 2026-08-13
"Marin is an open lab, in which the research and development of models is completely transparent from day 1"; each experiment tracked by a GitHub issue; Marin 8B Base trained for 12.7T tokens with a linked "full reproducible execution".
- https://huggingface.co/marin-community/marin-8b-base recorded 2026-08-13
Marin 8B Base model card, license apache-2.0, safetensors weight files, ungated.
- https://api.github.com/repos/marin-community/marin recorded 2026-08-13
`private: false`, `archived: false`, license `Apache-2.0`, description "Open-source framework for the research and development of foundation models", last pushed 2026-08-13.
Adoption
2 low confidence12,168 downloads in the trailing 30 days across the two declared artifacts (marin-community/marin-8b-base 8,604; marin-community/marin-8b-instruct 3,564), which bands at level 2 (10K-100K) on the model adoption scale. A research-stage fully-open project, on the map as a full-openness exemplar rather than for its popularity.
- https://huggingface.co/api/models/marin-community/marin-8b-base recorded 2026-08-13
8,604 downloads in the trailing 30 days for marin-community/marin-8b-base
- https://huggingface.co/api/models/marin-community/marin-8b-instruct recorded 2026-08-13
3,564 downloads in the trailing 30 days for marin-community/marin-8b-instruct
Capability
3 medium confidenceMid-tier capability for a fully-open model at 8B and 32B, a step below OLMo 3. Marin's own announcement puts 8B Base ahead of Llama 3.1 8B Base on 14 of 19 evaluations, and 8B Instruct ahead of OLMo 2 but short of Llama 3.1 Tulu, while Ai2's Olmo 3 post claims the strongest performance among fully open base models - so placing Marin one step under OLMo is what both vendors' own pages say.
- https://developers.googleblog.com/stanfords-marin-foundation-model-first-fully-open-model-developed-using-jax/ recorded 2026-08-13
Marin-8B-Base and Marin-8B-Instruct released with models, data, code and tokenizer under Apache 2.0; JAX/Levanter; the "Tootsie" process spanned over 12 trillion tokens
- http://marin.community/blog/2025/05/19/announcement/ recorded 2026-08-13
"On 14 out of 19 standard base model evals ... Marin 8B Base outperforms Llama 3.1 8B Base"; Marin 8B Instruct "outperforms OLMo 2 on standard instruct model evals, but still fall short of Llama 3.1 Tulu".
Verified 2026-08-13