Search Results for "video ocr"

...Despite being fairly lightweight (about 1 billion parameters), it delivers state-of-the-art performance across a wide variety of OCR tasks, outperforming many traditional OCR systems and even other multimodal models on benchmark suites. HunyuanOCR handles complex documents: multi-column layouts, tables, mathematical formulas, mixed languages, handwritten or stylized fonts, receipts, tickets, and even video-frame subtitles. The project provides code, pretrained weights, and inference instructions, making it feasible to deploy locally or on a server, and to integrate with applications.

MiniCPM-o 2.6 is a cutting-edge multimodal large language model (MLLM) designed for high-performance tasks across vision, speech, and video. Capable of running on end-side devices such as smartphones and tablets, it provides powerful features like real-time speech conversation, video understanding, and multimodal live streaming. With 8 billion parameters, MiniCPM-o 2.6 surpasses its predecessors in versatility and efficiency, making it one of the most robust models available. It supports...

Qwen3-Omni is a natively end-to-end multilingual omni-modal foundation model that processes text, images, audio, and video and delivers real-time streaming responses in text and natural speech. It uses a Thinker-Talker architecture with a Mixture-of-Experts (MoE) design, early text-first pretraining, and mixed multimodal training to support strong performance across all modalities without sacrificing text or image quality. The model supports 119 text languages, 19 speech input languages, and...