Two New OCR Models Land on Hugging Face: Tencent WeVisDoc and Jina OCR v1

Hugging Face 1 hour ago

Key Info

Two new OCR models were released on Hugging Face: Tencent's WeVisDoc (2B and 4B variants, Apache 2.0) and Jina AI's jina-ocr-v1 (non-commercial license). On the OmniDocBench v1.6 leaderboard, WeVisDoc-4B scores an overall 95.38, while the current SOTA remains NaviDC-OCR.

Highlights

Tencent/WeVisDoc comes in 2B and 4B sizes and is Apache 2.0 licensed, making it suitable for commercial use.
jinaai/jina-ocr-v1 is published under a non-commercial license.
On OmniDocBench v1.6, WeVisDoc-4B leads the new releases with 95.38 overall, but NaviDC-OCR is still the top model on the leaderboard.

Original post

RT @NielsRogge: 2 new OCR models landed on @huggingface today - tencent/WeVisDoc (2B and 4B), Apache 2.0 - jinaai/jina-ocr-v1, non-commercial license Which one should you pick? Based on the OmniDocBench v1.6 benchmark on Papers with Code, WeVisDoc-4B leads with an overall score of 95.38, but the current SOTA is NaviDC-OCR

#OCR#Hugging Face#WeVisDoc

View original on X