Can this be used for Jambo inference
freQuensy23-coder opened this issue · 1 comments
freQuensy23-coder commented
Can I use this solution for inference https://huggingface.co/ai21labs/Jamba-v0.1/discussions with offloading mamba moe layers?
Jambo it SOTA open source long context model and its support would be very useful for this library.
dvmazur commented
Hey, @freQuensy23-coder! The code in this repo is quite transformer-moe specific. I'm not too familiar with mamba-like architectures, but afaik @lavawolfiee has plans for adapting Jamba to work with our offloading strategy.