Starstruk Logo MarkStarstruk

v3.4.1

ggml-org/llama.cppv3.4.1Apr 14, 2026by Bobholamovic

AI Summary

This patch release focuses on expanding hardware support and fixing dependencies for PaddleOCR-VL. It adds `llama-cpp-server` backend compatibility, AMD and Intel Arc GPU support, resolves dependency issues for Huawei NPU and KunlunXin XPU, and removes page limits in the Docker Compose service.

Key Highlights

  • Added `llama-cpp-server` backend support for PaddleOCR-VL.
  • Added hardware support for AMD GPU and Intel Arc GPU.
  • Fixed dependency issues in images for Huawei NPU and KunlunXin XPU.
  • PaddleOCR-VL Docker Compose service no longer limits max pages per request.

New Features

  • llama-cpp-server backend support.
  • AMD GPU hardware support.
  • Intel Arc GPU hardware support.
  • Enhanced Docker Compose configuration (no page limit).

Full Release Notes

## 2026.4.14 v3.4.1 released

* PaddleOCR-VL adds `llama-cpp-server` backend support.
* PaddleOCR-VL adds AMD GPU and Intel Arc GPU hardware support.
* Fixed dependency issues in the PaddleOCR-VL images for Huawei NPU and KunlunXin XPU*
* For the PaddleOCR-VL Docker Compose service, the default configuration no longer limits the maximum number of pages per request.


## 2026.4.14 v3.4.1 发布

* PaddleOCR-VL 新增 `llama-cpp-server` 后端支持。
* PaddleOCR-VL 新增 AMD GPU、Intel Arc GPU 硬件支持。
* 修复 PaddleOCR-VL 华为 NPU、昆仑芯 XPU 镜像中的依赖问题。
* 对于 PaddleOCR-VL Docker Compose 服务，默认不限制单请求的页数上限。


**Full Changelog**: https://github.com/PaddlePaddle/PaddleOCR/compare/v3.4.0...v3.4.1