exo-explore
Enhancement: support local and custom models
At the moment, exo only supports loading models that are in the huggingface cache-folder. In case you have your own quants of a given model, you'd need to create a folder under the HF-cache that follows the conventions of HF-downloader (like for example: "~/.cache/huggingface/hub/models--mlx-community--Meta-Llama-3.1-8B-Instruct-4bit/refs/main") with a dummy commit sha hash.
Would be great if exo could support loading models from any given folder, to support local and custom models. Thanks!
Hi @vlbosch, @leo4life2,
I would love to take on this issue and work on supporting local and custom models in Exo! Please assign it to me.
Also, feel free to check out my portfolio at sanchay.me to learn more about my background and expertise.
I've also commented on two other issues that Iām interested in. If those get assigned elsewhere, I'd still love to work on this issue. Looking forward to contributing!
Best regards, Sanchay-T
