PaddleOCR 3.5: Running OCR and Document Parsing Tasks with a Transformers Backend
PaddleOCR 3.5 introduces support for running OCR and document parsing pipelines using a Hugging Face Transformers backend, enabling integration with the broader Transformers ecosystem. The update allows users to leverage transformer-based models for optical character recognition and structured document understanding tasks. This represents a convergence between the PaddlePaddle framework and the Transformers library for document AI workloads.
Related guides (3)
Related events (8)
PaddleOCR: OCR Toolkit Bridging Documents and LLMs
PaddleOCR is an open-source OCR toolkit built on PaddlePaddle that converts PDFs and images into structured data suitable for LLM pipelines. It supports 100+ languages and is positioned as a document-to-AI bridge. The repository has accumulated nearly 79,000 GitHub stars, with 148 new stars today, indicating sustained community interest.
Accelerating Document AI
This Hugging Face blog post covers the state of Document AI, focusing on tools and models for processing and understanding documents using machine learning. It likely discusses transformer-based approaches for tasks like document classification, information extraction, and visual document understanding. The post appears to survey the ecosystem of models and libraries available for document intelligence workflows.
Welcome PaddlePaddle to the Hugging Face Hub
Hugging Face announced the integration of PaddlePaddle, Baidu's open-source deep learning framework, into the Hugging Face Hub. This expands the Hub's ecosystem to support PaddlePaddle models alongside existing frameworks like PyTorch and TensorFlow. The move broadens access to Chinese-developed AI models and tooling within the broader ML community.
Accelerated Inference with Optimum and Transformers Pipelines
Hugging Face announced integration between the Optimum library and the Transformers Pipelines API, enabling hardware-accelerated inference with minimal code changes. The integration targets deployment on specialized hardware backends such as ONNX Runtime, allowing users to swap in optimized inference engines transparently. This lowers the barrier to production-grade inference optimization for practitioners using the Hugging Face ecosystem.
Transformers.js v3: WebGPU Support, New Models & Tasks, and More
Hugging Face released Transformers.js v3, a major update to its JavaScript inference library enabling on-device ML in browsers and Node.js. The release adds WebGPU backend support for hardware-accelerated inference, expands the supported model and task catalog, and improves overall performance. This brings browser-side AI inference closer to parity with native runtimes for a wider range of use cases.
Transformers v5: Simple model definitions powering the AI ecosystem
Hugging Face has announced Transformers v5, a major version update to its flagship open-source library. The release focuses on simplified model definitions and architectural improvements to the codebase. As one of the most widely used ML libraries in the ecosystem, this update has broad implications for researchers and practitioners building on top of the Transformers framework.
Mistral OCR 3: New Frontier in Document Processing Accuracy and Efficiency
Mistral AI has released Mistral OCR 3 (model ID: mistral-ocr-2512), claiming a 74% overall win rate over its predecessor Mistral OCR 2 across forms, scanned documents, complex tables, and handwriting. The model supports markdown output with HTML-based table reconstruction and is priced at $2 per 1,000 pages ($1 with Batch API). It now powers the Document AI Playground in Mistral AI Studio, offering a drag-and-drop interface for parsing PDFs and images into text or structured JSON.
Training a Language Model with Hugging Face Transformers Using TensorFlow and TPUs
This Hugging Face blog post provides a technical walkthrough for training a language model using TensorFlow and Google TPUs via the Transformers library. It covers the practical setup, data pipeline, and training configuration required to leverage TPU hardware with the TF ecosystem. The post serves as a tutorial bridging Hugging Face tooling with TPU-based infrastructure.