Loading
Open weights means a model's trained parameters are published so anyone can run and adapt it; that is less than open source, which also needs the training code and data, and more than gated weights, which require a request and a licence click.
Open-source AI, as the Open Source Initiative defines it, is an AI system freely available to use, study, modify and share, including its training data, code and parameters (Wikipedia). Most biology foundation models fall short of this: BulkFormer released weights and code under permissive terms; UNI2 and Virchow2 are gated on Hugging Face for non-commercial research; BulkRNABert is research-only; PLUTO and Med-Gemini released nothing. OnCo's open-source map records each project's openness level rather than treating open as one thing.
Shares Hugging Face Hub, Open licences: Apache-2.0, MIT and CC BY 4.0, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Open licences: Apache-2.0, MIT and CC BY 4.0, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Model cards and datasheets for datasets, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Pathology foundation models: UNI, UNI2, Virchow2, CTransPath, CONCH, TITAN, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Model cards and datasheets for datasets, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Local-only language models over patient data (privacy by architecture), Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Local-only language models over patient data (privacy by architecture), Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Model cards and datasheets for datasets, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.