b12246

<details open> vulkan: cap scalar flash attention at 8 rows for large heads on Qualcomm (#30829) With head sizes above 128 and KV >= 1024, scalar flash attention uses 16 rows per workgroup. On the Qualcomm Adreno driver this variant spills registers to scratch memory (264 bytes per invocation on the X1-85, none at 8 rows) and runs about 40x slower. The largest test-backend-ops case then becomes a single dispatch of several seconds, past the Windows GPU timeout, and the test never completes on the X1 Elite. Use at most 8 rows per workgroup on Qualcomm when hsk or hsv > 128. Smaller heads keep 16 rows, which is faster for them on the X2 Elite. Adreno X1-85, FLASH_ATTN_EXT hsk=256 kv=4096: nb=64 711 ms -> 17 ms, nb=512 no result -> 125 ms; test-backend-ops completes. Adreno X2-90: Gemma 4 E4B pp512 at depth 4096 with flash attention 81-90 -> 117 t/s. </details> **Website:** - <https://llama.app> **Attestations:** - <https://github.com/ggml-org/llama.cpp/attestations/65813913> **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-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/b12246/llama-b12246-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-arm64.tar.gz) - [Ubuntu 24.04 s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-s390x.tar.gz) - [Ubuntu 26.04 s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-26.04-s390x.tar.gz) - [Ubuntu 26.04 s390x (zDNN)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-zdnn-s390x.tar.gz) - [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-vulkan-x64.tar.gz) - [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-vulkan-arm64.tar.gz) - [Ubuntu x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-cuda-12.8-x64.tar.gz) - [CUDA 12.8 libraries](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-b12246-bin-ubuntu-cuda-12.8-x64.tar.gz) - [Ubuntu x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-cuda-13.5-x64.tar.gz) - [CUDA 13.5 libraries](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-b12246-bin-ubuntu-cuda-13.5-x64.tar.gz) - [Ubuntu arm64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-cuda-13.5-arm64.tar.gz) - [CUDA 13.5 libraries](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-b12246-bin-ubuntu-cuda-13.5-arm64.tar.gz) - [Ubuntu x64 (ROCm 10.0)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-rocm-10.0-x64.tar.gz) - [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-openvino-2026.4.1-x64.tar.gz) - [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-sycl-fp32-x64.tar.gz) - [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-ubuntu-sycl-fp16-x64.tar.gz) - [Linux arm64 (Snapdragon: CPU, Adreno GPU, Hexagon NPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-linux-arm64-snapdragon.tar.gz) - [setup guide](https://github.com/ggml-org/llama.cpp/blob/master/docs/backend/snapdragon/linux.md) **Android:** - [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-android-arm64.tar.gz) - [Android arm64 (Snapdragon: CPU, Adreno GPU, Hexagon NPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-android-arm64-snapdragon.tar.gz) - [setup guide](https://github.com/ggml-org/llama.cpp/blob/master/docs/backend/snapdragon/README.md) **Windows:** - [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-cpu-x64.zip) - [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-cpu-arm64.zip) - [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-opencl-adreno-arm64.zip) - [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-bin-win-cuda-12.4-x64.zip) - [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-cuda-13.5-x64.zip) - [CUDA 13.5 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-bin-win-cuda-13.5-x64.zip) - [Windows arm64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-cuda-13.5-arm64.zip) - [CUDA 13.5 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b12246/cudart-llama-bin-win-cuda-13.5-arm64.zip) - [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-vulkan-x64.zip) - [Windows arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-vulkan-arm64.zip) - [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-openvino-2026.4.1-x64.zip) - [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-sycl-x64.zip) - [Windows x64 (ROCm 10.0)](https://github.com/ggml-org/llama.cpp/releases/download/b12246/llama-b12246-bin-win-rocm-10.0-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/b12246/llama-b12246-ui.tar.gz)