Hugging Face has such a large repository of models! Curious how folks have been exploring HF models to date, and if this notebook-style environment to try out many models in one place would help?

I've tried running some using https://github.com/oobabooga/text-generation-webui on my arch box with an AMD gpu and it worked well and was really handy, until I tried yesterday and I get some python error no matter what I do, with the same environment as before and with a new one \o/

There doesn't seem to be an obvious easy way to run them with GPU support.

In current ML environment, there's no single thing as "GPU support" as nVidia and AMD are two very different (and unequal) worlds - if you want things to work on AMD, then you should look for things which are explicitly labeled as AMD-compatible, because the default state of "has GPU support" is slang for "works on CUDA, but also might work on AMD if stars are right".

Ah interesting! What models were you running inference on (assuming locally)?

I just ran the default one in the readme. My partner ran a few others from huggingface, but I don't know which.

Super interesting! Curious about the origin story of Lastmileai if you don’t mind sharing?

From the lastmile team here. Would love your feedback! Origin story -- we're a group of ML developers who want to build better developer tools to make AI more accessible to non-ML engineers. We built ML Ops tools in Big Tech for many years (Jupyter notebooks, model management, experimentation platforms, etc), but those were targeted towards ML researchers.

With LLMs and other generative AI models, we want to provide simpler tools to bridge that 'last mile' gap between the core foundation models, and using/personalizing/integrating them in your applications. Hence the name :)