Coding Tools

llama.cpp b11274 release with CUDA 13 and ROCm 10

llama.cpp ReleasesSeptember 30, 2026high confidence

Why it matters

→CUDA 13.4 support allows users to leverage the latest NVIDIA drivers for local inference.
→ROCm 10.0 integration improves compatibility with AMD GPUs on Linux and Windows.
→Snapdragon arm64 binaries expand hardware options for mobile and edge AI deployment.

llama.cpp has released build b11274, adding support for CUDA 13.4 and ROCm 10.0 across Linux and Windows platforms. The release also includes binaries for Linux arm64 targeting Snapdragon devices with CPU, Adreno GPU, and Hexagon NPU acceleration. Notably, KleidiAI support on macOS Apple Silicon is currently disabled in this build. The update ensures compatibility with the latest NVIDIA and AMD driver stacks without requiring manual compilation.

Read original

llama.cpp b11274 release with CUDA 13 and ROCm 10

Why it matters

More from llama.cpp Releases

llama.cpp fixes MoE dispatch inefficiency on Vulkan

llama.cpp Vulkan perf boost on Intel Arc

More in Coding Tools

Claude Code v2.1.284 adds Sonnet 5.5

Claude Code v2.1.285 patches security and stability

llama.cpp b11267 adds ROCm 10 and Snapdragon support

GitHub adds repo-level Dependabot runner config