b10443

ggml-org/llama.cppb10443Aug 15, 2026by github-actions[bot]same day as b10442

What changed since b10442

low risk

b10443 is a small hardening patch that validates GGUF array types before reading, with no other changes noted.

New features

  • GGUF loader now validates the array type field before reading array contents, adding a guard against malformed model files.

Improvements

  • More robust GGUF file parsing and safer handling of untrusted or corrupted model files.
  • No Vulkan/Intel Xe work is mentioned in this release; the prior release's backend optimizations carry over unchanged.

Bug fixes

  • Fixed a potential read error when loading GGUF files whose array type metadata is invalid or unexpected.

Upgrade notes

Drop-in replacement for b10442 with no config, CLI, or API changes mentioned. Rebuild or re-download binaries as usual; no model re-conversion is required. If you load GGUF files from untrusted sources, this build fails more cleanly on malformed array metadata.

Analyzed Oct 24, 2026

Upgrading?

Release notes summary

Bug fix release that adds validation for GGUF array types before reading, preventing potential errors with malformed files.

Key Highlights

  • Added array type validation
  • Prevents reading errors in GGUF files
  • Improved file handling robustness

Release Notes (excerpt)

<details open> fix: check gguf array type before reading (#27075) * fix: check gguf array type before reading * update skill </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 7.14)DISABLED - Ubuntu x64 (OpenVINO)…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories