Technology ❯ Artificial Intelligence ❯ Machine Learning

Optical Character Recognition

DeepSeek-OCR Document Parsing OCR Models OCR Datasets Text Recognition OCR Malware Data Privacy Sarvam Vision Document Processing AI in Software Development OCR Correction Pre-trained Models

DeepSeek-OCR Turns Text Into Images to Slash Context Tokens

The open release is drawing rapid experiments, with claims of steep compression facing questions about accuracy and reasoning.

DeepSeek Open-Sources OCR System That Compresses LLM Contexts With Visual Tokens

PaddlePaddle Releases PaddleOCR-VL, a 0.9B Vision-Language Model for Multilingual Document Parsing