Processing of high-resolution input with InternLM-XComposer2-4KHD

InternLM / InternLM-XComposer

InternLM-XComposer2 is a groundbreaking vision-language large model (VLLM) excelling in free-form text-image composition and comprehension.

1.91k stars 120 forks source link

Processing of high-resolution input with InternLM-XComposer2-4KHD #334

Open mariiak2021 opened 2 weeks ago

mariiak2021 commented 2 weeks ago

Hi,

I'm interested in the processing of high-resolution input with InternLM-XComposer2-4KHD (especially dynamic partitioning approach). Can you point for me please where is it happening in the code? Can't locate it myself unfortunatelly.

Best, Mariia