Hi @ruoqi-liu , is it possible for users to work with the vision encoder as a standalone model? Currently this is unclear based on documentation, and the demo (https://huggingface.co/spaces/aidhlab/PULSE-7B) reports a runtime error with an unknown error code (?) and no reason.
Thank you!
Hi @ruoqi-liu , is it possible for users to work with the vision encoder as a standalone model? Currently this is unclear based on documentation, and the demo (https://huggingface.co/spaces/aidhlab/PULSE-7B) reports a runtime error with an unknown error code (?) and no reason.
Thank you!