Coding Tools

llama.cpp b11056 adds CUDA 13 and ROCm 10.0

llama.cpp ReleasesSeptember 20, 2026high confidence

Why it matters

→CUDA 13 support removes the need for manual library management on newer NVIDIA stacks.
→ROCm 10.0 builds provide official parity for AMD GPU users on modern Linux distributions.
→Cross-platform availability ensures consistent inference capabilities across diverse hardware setups.

llama.cpp has released build b11056, expanding hardware compatibility with new binaries for CUDA 13 and ROCm 10.0. The update includes support for CUDA 12.8/13.3 on Linux and Windows, alongside ROCm 10.0 builds for both platforms. Apple Silicon KleidiAI is currently disabled in this version. The release also maintains standard CPU, Vulkan, and OpenVINO support across macOS, Linux, Windows, and Android.

Read original

llama.cpp b11056 adds CUDA 13 and ROCm 10.0

Why it matters

More from llama.cpp Releases

llama.cpp b11050 adds CUDA 13 and ROCm 10.0 support

llama.cpp b11052 adds CUDA 13 and ROCm 10 builds

More in Coding Tools

Claude Code v2.1.277 fixes critical bugs and adds AGENTS.md

Claude Code v2.1.278 optimizes auto-mode billing

llama.cpp b11054 adds Hexagon TOP_K support

GitHub Copilot Code Review GA with Auto-Resolution