Download Latest Version llama-b8611-bin-ubuntu-openvino-2026.0-x64.tar.gz (75.4 MB)
Email in envelope

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

Home / b8607
Name Modified Size InfoDownloads / Week
Parent folder
llama-b8607-xcframework.zip < 14 hours ago 174.2 MB
0
llama-b8607-bin-win-vulkan-x64.zip < 14 hours ago 56.1 MB
0
llama-b8607-bin-win-sycl-x64.zip < 14 hours ago 134.8 MB
0
llama-b8607-bin-win-opencl-adreno-arm64.zip < 14 hours ago 32.9 MB
0
llama-b8607-bin-win-hip-radeon-x64.zip < 14 hours ago 354.3 MB
0
llama-b8607-bin-win-cuda-13.1-x64.zip < 14 hours ago 158.4 MB
0
llama-b8607-bin-win-cuda-12.4-x64.zip < 14 hours ago 232.7 MB
0
llama-b8607-bin-win-cpu-x64.zip < 14 hours ago 38.8 MB
0
llama-b8607-bin-win-cpu-arm64.zip < 14 hours ago 31.7 MB
0
llama-b8607-bin-ubuntu-x64.tar.gz < 14 hours ago 31.3 MB
0
llama-b8607-bin-ubuntu-vulkan-x64.tar.gz < 14 hours ago 48.5 MB
0
llama-b8607-bin-ubuntu-vulkan-arm64.tar.gz < 14 hours ago 40.7 MB
0
llama-b8607-bin-ubuntu-s390x.tar.gz < 14 hours ago 34.8 MB
0
llama-b8607-bin-ubuntu-rocm-7.2-x64.tar.gz < 14 hours ago 153.3 MB
0
llama-b8607-bin-ubuntu-openvino-2026.0-x64.tar.gz < 14 hours ago 75.4 MB
0
llama-b8607-bin-ubuntu-arm64.tar.gz < 14 hours ago 27.6 MB
0
llama-b8607-bin-macos-x64.tar.gz < 14 hours ago 101.5 MB
0
llama-b8607-bin-macos-arm64.tar.gz < 14 hours ago 39.9 MB
0
llama-b8607-bin-910b-openEuler-x86-aclgraph.tar.gz < 14 hours ago 71.1 MB
0
llama-b8607-bin-910b-openEuler-aarch64-aclgraph.tar.gz < 14 hours ago 63.7 MB
0
llama-b8607-bin-310p-openEuler-x86.tar.gz < 14 hours ago 71.1 MB
0
llama-b8607-bin-310p-openEuler-aarch64.tar.gz < 14 hours ago 63.7 MB
0
cudart-llama-bin-win-cuda-13.1-x64.zip < 14 hours ago 402.6 MB
0
cudart-llama-bin-win-cuda-12.4-x64.zip < 14 hours ago 391.4 MB
0
b8607 source code.tar.gz < 19 hours ago 29.6 MB
0
b8607 source code.zip < 19 hours ago 30.8 MB
0
README.md < 19 hours ago 3.3 kB
0
Totals: 27 Items   2.9 GB 0
ggml webgpu: quantized buffers to u32 + wider browser/device support (#21046) * Work towards removing bitcast * Move rest of existing types over * Add timeout back to wait and remove synchronous set_tensor/memset_tensor * move to unpackf16 for wider compatibility * cleanup * Remove deadlock condition in free_bufs

macOS/iOS:

Linux:

Windows:

openEuler:

Source: README.md, updated 2026-04-01