llava-gguf / README.md
calcuis's picture
Update README.md
7e54df7 verified
|
raw
history blame
440 Bytes
metadata
license: apache-2.0
language:
  - en
pipeline_tag: image-text-to-text
tags:
  - llava
  - vision

llava: computer vision GGUF

  • Clip Handler > mmproj-***.gguf
  • Vision Model > llava-***.gguf
  • provide a link of picture as prompt
  • it will give hints for you to comply your text-to-image prompt (reverse engineering) 🎃

how to use

  • compatible with gguf-connector (next upgrade will work)
  • work in progress (coming in few hours)