b10204

ggml-org/llama.cppb10204Jul 31, 2026by github-actions[bot]same day as b10203

What changed since b10203

low risk

b10204 adds SYCL device-to-device memcpy forwarding, an Intel-focused optimization on top of b10203's Q2 SYCL matmul support.

New features

  • SYCL backend now supports device-to-device memory copies via the DEV2DEV_MEMCPY_FORWARD mechanism.

Improvements

  • Memory transfers between SYCL devices can be forwarded directly, reducing overhead on Intel hardware.

Upgrade notes

This is a single-commit increment over b10203 with no API, config, or model-format changes. SYCL users on Intel hardware get the benefit automatically; other backends are unaffected. No migration steps required.

Analyzed Oct 31, 2026

Upgrading?

Release notes summary

This release adds support for device-to-device memory copying in SYCL using the DEV2DEV_MEMCPY_FORWARD mechanism, optimizing performance for Intel hardware.

Key Highlights

  • SYCL support for dev2dev memcpy
  • Intel-specific optimization for memory transfers

Release Notes (excerpt)

<details open> sycl : support dev2dev memcpy by DEV2DEV_MEMCPY_FORWARD (#26234) Co-authored-by: Neo Zhang Jianyu <jianyu.zhang@intel.com> </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)…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories