marin
A Stanford-originated open research program that builds large language models entirely in public, sharing every dataset, training run, evaluation and failed experiment so others can reproduce and build on the work.
View marin-community/marin on GitHub- GitHub stars
- 4,098
- Forks
- —
- Language
- Python
Project details
- Repository
- marin-community/marin
- Licence
- Apache-2.0
- Primary language
- Python
- Catalog status
- Available
- Catalog check
- 2026-11-13
Repository description
Open-source framework for the research and development of foundation models.
Explore by topic
More tools in Apps & utilities
- spec-kit
github · MIT
GitHub own open-source toolkit for Spec-Driven Development: you write a clear spec first, and any AI coding agent builds the implementation from it.
- nanochat
karpathy · MIT
nanochat is Andrej Karpathy full, minimal recipe to train and chat with your own ChatGPT-style model for about $100 in compute.
- kronos
shiyu-coder · MIT
An open-source AI foundation model trained specifically to model and predict patterns in financial market data.
- financial-services
anthropics · Apache-2.0
Anthropic open-sourced a set of Claude agents that draft real finance work: investment memos, research notes, reconciliations, for a human to review and sign off on.