Starstruk Logo MarkStarstruk

b10517

ggml-org/llama.cppb10517Aug 20, 2026by github-actions[bot]

AI Summary

Optimizes Flash Attention in the Vulkan backend by dequantizing KV cache only once in cooperative matrix operations and handling edge cases like FA scratch buffer limits.

Key Highlights

  • Optimized KV dequantization in Vulkan Flash Attention
  • Added fallback for FA scratch buffer exceeding maxStorageBufferRange
  • Added tests for contiguous KV allocation

New Features

  • Vulkan Flash Attention KV optimization

Full Release Notes

<details open>

vulkan : dequant q8_0 KV once in coopmat1 (#25494)

* vulkan : dequant q8_0 KV once in coopmat1

Assisted-by: Claude (Opus 4.8)

* vulkan : fall back instead of aborting when FA scratch exceeds maxStorageBufferRange

* vulkan : require KV-cache layout in FA dequant path

Assisted-by: Claude (Opus 4.8)

* vulkan : skip FA dequant path on coopmat2

Assisted-by: Claude (Opus 4.8)

* tests : add contiguously-allocated quant K/V FA tests

Assisted-by: Claude (Opus 4.8)

* vulkan : trim comments

* vulkan : tighten permutation checks for FA path

* vulkan : set prealloc_x_need_sync after the FA dispatch

* vulkan : exclude Intel Xe1 from FA dequant path

</details>

**Website:**
- <https://llama.app>

**Attestations:**
- <https://github.com/ggml-org/llama.cpp/attestations/41936012>

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-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/b10517/llama-b10517-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-vulkan-arm64.tar.gz)
- Ubuntu x64 (ROCm 7.14)[DISABLED](https://github.com/ggml-org/llama.cpp/pull/26969)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-openvino-2026.3-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10517/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10517/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows arm64 (CUDA 13) (preview)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-cuda-13.4-arm64.zip) - [CUDA 13.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10517/cudart-llama-bin-win-cuda-13.4-arm64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-openvino-2026.3-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-sycl-x64.zip)
- [Windows x64 (ROCm 7.14)](https://github.com/ggml-org/llama.cpp/releases/download/b10517/llama-b10517-bin-win-rocm-7.14-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/b10517/llama-b10517-ui.tar.gz)