Download Latest Version llama-b7347-bin-ubuntu-s390x.tar.gz (18.5 MB)
Email in envelope

Get an email when there's a new version of llama.cpp

Home / b7340
Name Modified Size InfoDownloads / Week
Parent folder
llama-b7340-xcframework.zip < 24 hours ago 147.1 MB
llama-b7340-xcframework.tar.gz < 24 hours ago 147.2 MB
llama-b7340-bin-win-vulkan-x64.zip < 24 hours ago 31.9 MB
llama-b7340-bin-win-sycl-x64.zip < 24 hours ago 106.1 MB
llama-b7340-bin-win-opencl-adreno-arm64.zip < 24 hours ago 14.1 MB
llama-b7340-bin-win-hip-radeon-x64.zip < 24 hours ago 343.3 MB
llama-b7340-bin-win-cuda-13.1-x64.zip < 24 hours ago 89.9 MB
llama-b7340-bin-win-cuda-12.4-x64.zip < 24 hours ago 200.6 MB
llama-b7340-bin-win-cpu-x64.zip < 24 hours ago 17.0 MB
llama-b7340-bin-win-cpu-arm64.zip < 24 hours ago 13.6 MB
llama-b7340-bin-ubuntu-x64.zip < 24 hours ago 16.1 MB
llama-b7340-bin-ubuntu-x64.tar.gz < 24 hours ago 16.1 MB
llama-b7340-bin-ubuntu-vulkan-x64.zip < 24 hours ago 31.5 MB
llama-b7340-bin-ubuntu-vulkan-x64.tar.gz < 24 hours ago 31.5 MB
llama-b7340-bin-ubuntu-s390x.zip < 24 hours ago 15.8 MB
llama-b7340-bin-ubuntu-s390x.tar.gz < 24 hours ago 18.6 MB
llama-b7340-bin-macos-x64.zip < 24 hours ago 38.2 MB
llama-b7340-bin-macos-x64.tar.gz < 24 hours ago 38.2 MB
llama-b7340-bin-macos-arm64.tar.gz < 24 hours ago 13.9 MB
llama-b7340-bin-macos-arm64.zip < 24 hours ago 13.9 MB
cudart-llama-bin-win-cuda-13.1-x64.zip < 24 hours ago 402.6 MB
cudart-llama-bin-win-cuda-12.4-x64.zip < 24 hours ago 391.4 MB
b7340 source code.tar.gz 2025-12-09 28.1 MB
b7340 source code.zip 2025-12-09 29.0 MB
README.md 2025-12-09 3.4 kB
Totals: 25 Items   2.2 GB 0

[!WARNING] Release Format Update: Linux releases will soon use .tar.gz archives instead of .zip. Please make the necessary changes to your deployment scripts.

metal: SSM kernel improvements (#17876) * feat: Add a batched version of ssm_conv This was done using Claude Code. It found a number of optimizations around how the threads were organized, resulting in a huge performance boost! Branch: Mamba2SSD Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: Optimized SSM_SCAN kernel for metal This used Claude Code and resulted in a modest performance improvement while maintaining correctness. Branch: Mamba2SSD Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * test: Add test-backend-ops perf tests for SSM_CONV Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * test: Real representitive tests for SSM_CONV Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: Use function constant for ssm_conv batch size Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * test: backend op tests for ssm_scan from granite4 1b-h Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * style: remove commented out templates Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: float4 version of ssm_conv_batched Branch: SSMKernelImprovements Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Add missing ggml_metal_cv_free Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

macOS/iOS: - macOS Apple Silicon (arm64) - macOS Intel (x64) - iOS XCFramework

Linux: - Ubuntu x64 (CPU) - Ubuntu x64 (Vulkan) - Ubuntu s390x (CPU)

Windows: - Windows x64 (CPU) - Windows arm64 (CPU) - Windows x64 (CUDA 12) - Windows x64 (CUDA 13) - Windows x64 (Vulkan) - Windows x64 (SYCL) - Windows x64 (HIP)

Source: README.md, updated 2025-12-09