DeepSeek OCR 2 is an open-source OCR and document understanding model built for complex layouts, Markdown output, and human-like reading order.

• #Document OCR
• #Document to Markdown
• #Image to Text

DeepSeek OCR is an open-source vision-language OCR model that converts document images into structured text and Markdown with efficient visual token compression.

• #Document OCR
• #Document to Markdown
• #Image to Text

LTX-2 is an open-source AI video model that generates synchronized video and audio for creative, research, and production workflows.

VoxCPM is an open-source TTS model family for multilingual speech generation, voice design, and realistic voice cloning.

Breakthrough in Emotionally Expressive and Duration-Controlled Auto-Regressive Zero-Shot Text-to-Speech