LLaVA-OneVision: Easy Visual Task Transfer
Paper
•
2408.03326
•
Published
•
61
LLaVA-OneVision
a model good at arbitrary types of visual input
-
-
lmms-lab/LLaVA-OneVision-Mid-Data
Viewer • Updated • 563k • 608 • 20 -
lmms-lab/LLaVA-OneVision-Data
Viewer • Updated • 3.94M • 29.1k • 196 -
lmms-lab/LLaVA-NeXT-Data
Viewer • Updated • 779k • 3.14k • 36 -
lmms-lab/llavanext-qwen-siglip-tokenizer
Text Generation • Updated • 17 • 3 -
lmms-lab/llava-onevision-qwen2-0.5b-si
Text Generation • 0.9B • Updated • 4.22k • 13 -
lmms-lab/llava-onevision-qwen2-0.5b-ov
Text Generation • 0.9B • Updated • 31.1k • 19 -
lmms-lab/llava-onevision-qwen2-7b-si
Text Generation • 8B • Updated • 2.19k • 12 -
lmms-lab/llava-onevision-qwen2-7b-ov
Text Generation • 8B • Updated • 66.5k • 52 -
lmms-lab/llava-onevision-qwen2-72b-si
Text Generation • 73B • Updated • 72 • 1 -
lmms-lab/llava-onevision-qwen2-72b-ov-sft
Text Generation • 73B • Updated • 392 • 14 -
lmms-lab/llava-onevision-qwen2-72b-ov-chat
Image-Text-to-Text • 73B • Updated • 643 • 9 -
lmms-lab/llava-onevision-projectors
Updated • 3 -
lmms-lab/llava-onevision-qwen2-0.5b-mid-stage-a4
1B • Updated • 16 -
lmms-lab/llava-onevision-qwen2-7b-mid-stage-a4
8B • Updated • 21