Multimodal model

GOT-OCR2_0

Listed for image text to text. Explore the model card for usage and requirements.

Type
Multimodal
Weights
Open
Downloads
603k
Likes
1.6k
Pipeline
image text to text
License
apache-2.0

Model hub

stepfun-ai/GOT-OCR2_0

Repo
stepfun-ai/GOT-OCR2_0
Access
Public
Technical tags and languages
safetensors GOT got vision-language ocr2.0 custom_code image-text-to-text multilingual
Open on Hugging Face

Papers

3

More from stepfun-ai

1