how to run the model locally?
#7
by
828CFXLpyz
- opened
Expecting I can query it with curl.
To clarify, do you mean actually downloading the model and running inference? Or querying an API to run inference?
For the former, the instructions are in the README.
For the latter, the easiest way is to follow the instructions given by the link "Use via API" in the footer of the demo. This will let you run inference locally through the Gradio API. However, this is not meant for production use and is not performant: its running on a small GPU without all optimizations and your requests will be queued.
Hope that helps!
pirroh
changed discussion status to
closed