About the acceleration frame work.

InternLM / InternLM-XComposer

InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output

2.06k stars 128 forks source link

About the acceleration frame work. #210

Open Yxxxb opened 3 months ago

Yxxxb commented 3 months ago

Hi!

I wonder if you have tried to deploy any acceleration or memory saving framework for InternLM-XComposer? like vLLM. If so can you provide code or possible solution?

Thanks!