llama.cpp b9559 fixes prompt spinner, expands platform builds
A tiny UI fix and 20+ platform builds land in the latest release.
The open-source llama.cpp project, maintained by ggml-org, released version b9559 on June 8. The primary change is a fix for the CLI spinner not showing during prompt processing (issue #24283), a minor UI quality-of-life improvement for users running text generation from the terminal.
Notably, the release expands build coverage across 20+ platform/target combinations. This includes macOS Apple Silicon (both standard and KleidiAI-enabled), macOS Intel, iOS XCFramework, Linux on x64, arm64, s390x (CPU and Vulkan), plus specialized Linux builds for ROCm 7.2, OpenVINO, and SYCL. Windows users get x64 and arm64 CPU builds, CUDA 12 and CUDA 13 DLL variants, Vulkan, SYCL, and HIP. An Android arm64 build and openEuler builds (with ACL Graph for 310p/910b) are also included, though several are marked DISABLED. The UI assets numbered 23 suggest no major visual changes beyond the spinner fix.
- Fixes CLI spinner not appearing during prompt processing (issue #24283).
- Ships prebuilt binaries for macOS, Windows, Linux, Android, iOS, and openEuler across 20+ configurations.
- No new model support, performance gains, or architectural changes in this point release.
Why It Matters
A minor fix but reinforces llama.cpp's commitment to polish across many hardware backends professionals use.