
Academic lab · Stanford, United States
Stanford CRFM
Stanford's Center for Research on Foundation Models, creators of the HELM evaluation framework and Alpaca.
Stanford, United States
- Founded
- 2021
- Type
- Academic lab
- Based in
- Stanford, United States
- Key papers
- 3
- Featured projects
- 7
About
CRFM is a Stanford HAI center directed by Percy Liang. It studies foundation models and builds open tools for evaluating them.
HELM, published in 2022, benchmarked 30 models on 42 scenarios and 7 metrics under standard conditions, and is maintained as a living benchmark. The center also produced Alpaca, a fine-tuned LLaMA 7B model.
It releases BioMedLM and training code on GitHub and Hugging Face.
Mission and ambitionsCRFM aims to make foundation models more transparent and understood through open evaluation and research.
Known for
- 01
HELM evaluation
- 02
Alpaca
- 03
BioMedLM
- 04
Foundation model research
Milestones
- 2021
CRFM founded
Stanford HAI center for research on foundation models established.
- 2021
On the Opportunities and Risks of Foundation Models
Report coining the term and surveying the field.
- 2022
HELM
Published Holistic Evaluation of Language Models.
- 2023
Alpaca
Released an instruction-tuned LLaMA 7B model.
- 2023
Foundation Model Transparency Index
Scored major developers on disclosure.
Open source
Holistic Evaluation of Language Models.
Biomedical LM.
Framework for transparent GPT-2 training.
Foundation model ecosystem data.
- air-bench-2024dataset
Safety benchmark.
- image2struct-latex-v1dataset
Image-to-structure benchmark.
- music-small-800kmodel
Music model.
Research papers
Notable releases
Openness
Open code and models.
People
- Percy LiangDirector
Associate professor of computer science; leads HELM.
- Rishi BommasaniSociety lead and researcher
Lead author of the foundation models report and the Foundation Model Transparency Index.
- Tatsunori HashimotoFaculty researcher
Co-led the Alpaca project.
Listed from the lab's own pages and public profiles.
Contact and community
Contact details and links are on the lab's website (https://crfm.stanford.edu); code questions go through issues on their GitHub organisation.
Sources
Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.



