b10075

ggml-org/llama.cppb10075Jul 20, 2026by github-actions[bot]1 days after b10069

What changed since b10069

low risk

Delta is a single Hexagon backend addition (CLAMP op); intermediate builds b10070–b10074 are undocumented.

New features

  • Hexagon backend now supports the CLAMP operation, widening the set of graphs that can run on Hexagon GPUs.

Improvements

  • Expanded Hexagon offload coverage reduces fallback of clamp-containing graphs to CPU.

Bug fixes

  • Resolves issue #25934 by enabling CLAMP on Hexagon instead of failing or falling back.

Upgrade notes

No config or API changes are indicated; Hexagon users simply gain CLAMP support. Note that the release notes only document one commit, so builds b10070–b10074 may contain unlisted changes. Rebuild or re-download binaries if you rely on Hexagon offload.

Analyzed Oct 24, 2026

Upgrading?

Release notes summary

Hexagon backend adds CLAMP operation support.

Key Highlights

  • Add CLAMP op to Hexagon backend
  • Extends Hexagon GPU support for more operations
  • Fixes issue #25934

Release Notes (excerpt)

<details open> hexagon: add CLAMP op (#25934) </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:** - Android arm64 (CPU)…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories