4 ms·
Where I'm struggling at the moment is that I know about those but my local hardware is a bit limited and I haven't figured out how the dots connect between runn
by indigochill 3y ago
Where I'm struggling at the moment is that I know about those but my local hardware is a bit limited and I haven't figured out how the dots connect between running those local interfaces against (affordable) rented GPU servers. The info I can find assumes you're running everything locally.
For example, I know HuggingFace provides inference endpoints, but I haven't found information for how to connect Oobabooga to those endpoints. The information's probably out there. I just haven't found it yet.
- amelius 3y agoWhere I'm struggling is how to keep up to date on the latest LLMs and their performance.
- bioemerl 3y agoThere is something called a run pod but I know I've seen a couple of these groups give quick easy links to use. You might want to look there. > I know HuggingFace provides inference endpoints, but I haven't found information for how to connect Oobabooga to those endpoints I've never heard of these so I'm guessing there isn't a way.