🐣 Please follow me for new updates https://twitter.com/camenduru
🔥 Please join our discord server https://discord.gg/k5BwmmvJJU
🥳 Please join my patreon community https://patreon.com/camenduru

🦒 Colab

Colab	Info
	🌋 LLaVA_13b_4bit_colab 13B (4bit)
	🌋LLaVA_13b_4bit_vanilla_colab 13B (4bit) (without gradio)
	🌋LLaVA_13b_4bit_caption_colab 13B (4bit) (without gradio - image caption loop)
	🌋 LLaVA_7b_8bit_colab 7B (8bit)
	🌋 LLaVA_7b_colab 7B (16bit) (Pro High-RAM 😐 22GB RAM 14GB VRAM)

⚠ Tutorial

https://www.youtube.com/watch?v=o7zQAa0NPds

After clicking the third cell, please wait for the model to load.
~14.7GB for 16bit ~8GB for 8bit ~5 minutes
We will not receive any output because it will be running in another thread.

Main Repo

https://github.com/haotian-liu/LLaVA

Paper

https://arxiv.org/abs/2304.08485

Page

https://llava-vl.github.io/

Output