b10256

ggml-org/llama.cppb10256Aug 4, 2026by github-actions[bot]

AI Summary

Optimizes SYCL performance by parallelizing the non-contiguous concatenation kernel.

Key Highlights

  • Parallelized non-contiguous concat kernel
  • Performance improvement (~9.4%)
  • Fixed block size cap

New Features

  • SYCL kernel parallelization

Full Release Notes

<details open>

sycl: parallelize the non-contiguous concat kernel (#25852)

* sycl: parallelize the non-contiguous concat kernel

Launch geometry only: the non-contiguous concat kernel launched a single-lane
work-group (1, 1, 1), now it will launch a (1, 1, SYCL_CONCAT_BLOCK_SIZE) one.

SYCL_CONCAT_BLOCK_SIZE is defined in `ggml/src/ggml-sycl/presets.hpp`.

llama-bench (Arc Pro B70, Qwen3.6-27B-UD-Q4_K_XL, -fa on, q8_0 KV),
on top of upstream master: pp2048 920 -> 1006 t/s (+9.4%)

* sycl: cap non-contiguous concat block at ne0

* sycl: make non-contiguous concat block width env-tunable (GGML_SYCL_CONCAT_BLOCK_SIZE)

* Revert "sycl: make non-contiguous concat block width env-tunable (GGML_SYCL_CONCAT_BLOCK_SIZE)"

This reverts commit 2709909e790d298cc0746950caff16ad8ecf9d6f.

</details>

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

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-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/b10256/llama-b10256-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-vulkan-arm64.tar.gz)
- [Ubuntu x64 (ROCm 7.2)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-rocm-7.2-x64.tar.gz)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-openvino-2026.2.1-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10256/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b10256/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-openvino-2026.2.1-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-bin-win-sycl-x64.zip)
- [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/b10256/llama-b10256-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/b10256/llama-b10256-ui.tar.gz)