b10151

ggml-org/llama.cppb10151Jul 27, 2026by github-actions[bot]same day as b10150

What changed since b10150

low risk

b10151 only parallelizes SYCL ocloc build steps; no runtime or API changes versus b10150.

Improvements

  • SYCL builds now invoke ocloc in parallel, cutting OpenCL kernel compilation time.
  • Faster CI/build turnaround for SYCL-enabled binaries; runtime behavior is unchanged.

Upgrade notes

This is a build-system-only change, so no config, model, or API migration is needed. Users of prebuilt SYCL packages get identical runtime behavior; only those compiling from source will notice faster builds. The earlier b10150 offloading-logic and DeepSeek-V4 graph changes remain in effect.

Analyzed Nov 1, 2026

Upgrading?

Release notes summary

Parallelized the OpenCL compiler (ocloc) invocations in the SYCL build process for better build performance.

Key Highlights

  • Parallelized SYCL build invocations for improved compilation speed

New Features

  • Parallelized SYCL ocloc build invocations

Release Notes (excerpt)

<details open> sycl(build): parallelize ocloc invocations (#25903) </details> **Website:** - <https://llama.app> **macOS/iOS:** - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED - macOS Intel (x64) - iOS XCFramework **Linux:** - Ubuntu x64 (CPU) - Ubuntu arm64 (CPU) - Ubuntu s390x (CPU) - Ubuntu x64 (Vulkan) - Ubuntu arm64 (Vulkan) - Ubuntu x64 (ROCm 7.2) - Ubuntu x64 (OpenVINO) - Ubuntu x64 (SYCL FP32) - Ubuntu x64 (SYCL FP16) **Android:**…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories