Promote Sign in Sign up

Text model

Jina-VLM

Jina-VLM is a 2.4B-parameter small multilingual vision-language model that links a SigLIP2 vision encoder with a Qwen3 backbone to deliver strong multilingual visual question answering on images.

Jina AI Text Open weights cc-by-nc-4.0
Visit model website ↗ Hugging Face
Type
Text
Weights
Open
Downloads
555
Likes
124
Pipeline
image text to text
License
cc-by-nc-4.0

Model hub

jinaai/jina-vlm

Repo
jinaai/jina-vlm
Library
transformers
Access
Public
Technical tags and languages
en zh ar pt ru tr de es fr it ja transformers safetensors jvlm text-generation multimodal multilingual vlm vision-language qwen3 siglip2 image-text-to-text conversational
Open on Hugging Face

More from Jina AI

4

About Jina AI

1

Jina AI is a company that specializes in AI technology with a focus on multimodal AI, offering open-source technology and innovative solutions.