Biomni/Biomni-R0-32B-Preview is a 32 billion parameter research preview model developed by the Biomni team at Stanford. This model is trained with end-to-end reinforcement learning using the Biomni-E1 environment scaffolding. It is specifically designed for biomedical AI agent tasks, achieving state-of-the-art performance across ten diverse biomedical benchmarks including CRISPR delivery and rare disease diagnosis. The model supports a context length of 32768 tokens, with options for extended context via rope scaling.
No reviews yet. Be the first to review!