OctoThinker

/

OctoThinker-3B-Long-Zero

like 0

Follow

OctoThinker 32

This PR adds a model card, linking the model to the paper OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling as well as adding the relevant metadata (license, pipeline tag, library name) for more discoverability.
Project page: https://huggingface.co/OctoThinker.