Models & Labs
PP-OCRv6 Launches on Hugging Face with 50-Language Support
Hugging Face Bloghigh confidence
Why it matters
- →PP-OCRv6 enhances OCR accuracy while maintaining lightweight models for flexible deployment.
- →It supports 50 languages, reducing the need for multiple OCR models in multilingual scenarios.
- →The integration with various backends like PaddlePaddle and Transformers broadens its applicability.
PP-OCRv6, the latest iteration of PaddleOCR's model family, is now available on Hugging Face. This release supports 50 languages and offers models ranging from 1.5M to 34.5M parameters. It improves text detection and recognition accuracy over its predecessor, PP-OCRv5_server. The models are designed for real-world applications, with flexible deployment options across PaddlePaddle, Transformers, and ONNX Runtime. This makes PP-OCRv6 a versatile choice for developers needing robust OCR solutions in multilingual contexts.
Read originalMore in Models & Labs
Models & Labsmodels
llama.cpp b10955 release addresses heap corruption
The latest release of llama.cpp, b10955, tackles a critical issue of heap corruption by disabling the ggml-cpu precompiled header and fixing CACHE_LINE_SIZE ambiguity. This update ensures consistent CACHE_LINE_SIZE values across C++ kernels and C work-buffer sizing code, preventing heap-buffer-overflow and subsequent crashes. By restoring the natural include order and removing the std::hardware_destructive_interference_size branch, the update makes the value deterministic and include-order independent. This release is a technical fix that stabilizes the runtime environment for developers using llama.cpp.
llama.cpp Releases
Models & Labsmodels
llama.cpp b10956 release enhances SYCL backend
The latest llama.cpp release, b10956, introduces significant improvements to the SYCL backend, particularly for handling large k values in TOP_K operations. By implementing a radix select method, the update allows for efficient GPU-resident processing, avoiding previous limitations that forced operations to fall back to the CPU. This change enhances performance, especially in scenarios requiring large k values, such as qwen4exp's sparse-attention indexer. The update ensures that operations are more efficient and scalable, providing a notable boost in processing speed without regressing any measured shapes.