Open byjlw opened 1 week ago
Hi,
In the v0.2.0 release we'll add support for direct ExecuTorch bindings. This will allow users to load any model as long as it's exported to .pte
. When that comes out, we'll add a quick runnable example that shows how to use the API. For now, the useLLM
hook is described in the docs.
Hi @chmjkb
Are you planning to add support for other uses cases like super resolution or image transfer?
Cause right now the useLLM
only allows a text in the generate
function right?
It would be awesome if we could have a quick start runnable example in the main readme. Nothing major but start to finish on how to get a model wire it to the react framework and run it to interact with the model. Ideally less than 5 steps.