b9828

ggml-org/llama.cppb9828Jun 27, 2026by github-actions[bot]

AI Summary

Significant performance and feature updates to the OpenCL backend including a major rework of Flash Attention kernels supporting various quantizations and precisions.

Key Highlights

  • Reworked Flash Attention kernels for f16/f32
  • Added prefill prepass kernels
  • Added support for q4_0 and q8_0 quantizations
  • Added set_rows functionality for f32 to q8_0/q4_0

New Features

  • OpenCL Flash Attention rework (f16/f32)
  • OpenCL Flash Attention support for q4_0 and q8_0
  • OpenCL prefill prepass kernels
  • Dequant kernels for q4_0 and q8_0

Full Release Notes

<details open>

opencl: flash attention improvement (#25069)

* opencl: rework FA kernel for f16 and f32

* opencl: flash-attention prefill prepass kernels

- flash_attn_kv_pad_f16    pads the tail KV tile to a BLOCK_N multiple
- flash_attn_mask_pad_f16  pads the matching mask tile
- flash_attn_blk_f16       classifies each KV tile per query block as
                           fully masked / mixed / fully unmasked, so
                           the main kernel can skip fully-masked tiles
                           and the mask lookup for fully-unmasked ones

* opencl: FA kernels for q4_0 and q8_0

* opencl: `set_rows` for f32 to q8_0/q4_0

* opencl: dequant kernels for q4_0 and q8_0

* opencl: add FA tile tuning table with override

* opencl: wire host side for FA

* opencl: q4_0 MoE tensors are also SOA'ed

* opencl: cosmetic fix

* opencl: refactor, also clarify some code paths in comments

* opencl: fix inifity for `-cl-finite-math-only`

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>

</details>

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-macos-arm64.tar.gz)
- macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780)
- [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-vulkan-arm64.tar.gz)
- [Ubuntu x64 (ROCm 7.2)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-rocm-7.2-x64.tar.gz)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-openvino-2026.2.1-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9828/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9828/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-openvino-2026.2.1-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-sycl-x64.zip)
- [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-bin-win-hip-radeon-x64.zip)

**openEuler:**
- [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23705)
- openEuler x86 (310p)
- openEuler x86 (910b, ACL Graph)
- openEuler aarch64 (310p)
- openEuler aarch64 (910b, ACL Graph)

**UI:**
- [UI](https://github.com/ggml-org/llama.cpp/releases/download/b9828/llama-b9828-ui.tar.gz)