Notice: Upcoming Default Server Port Change
The latest llama.cpp release (build b10240) introduces a notice about a planned change to the default server port. Currently set to 8080, the port will move to 9931 in a future update. Self-hosted users who rely on the default configuration should be aware that scripts, clients, and firewall rules may need adjustment once the change takes effect. This heads-up gives you time to prepare your setups.
Pre-built Binaries for All Major Platforms
Alongside the notice, b10240 provides an extensive set of pre-built binaries, making it easy to run llama.cpp on your hardware. Here is what is available:
- macOS: Apple Silicon (arm64) and Intel (x64). (KleidiAI enabled build for arm64 is currently disabled.)
- iOS: XCFramework.
- Linux (Ubuntu): x64 and arm64 variants for CPU, Vulkan, ROCm 7.2, OpenVINO, and SYCL (FP32 and FP16). Additionally, an s390x CPU build is available.
- Android: arm64 CPU.
- Windows: x64 CPU, arm64 CPU, OpenCL Adreno (arm64), CUDA 12 (with DLLs), CUDA 13 (with DLLs), Vulkan, OpenVINO, SYCL, and HIP (Radeon).
- openEuler: x86 and aarch64 builds for 310p and 910b (ACL Graph). Note: A related pull request link is currently disabled.
A standalone UI package is also available for those who prefer a graphical interface. For more information, visit llama.app.
