Curious as to why the text encoders are included in this repo. Are they fine-tuned as well, or can we use the original llama and CLIP models?
Text encoders are the same.
Great thanks @PY007 .
· Sign up or log in to comment
