b9964

ggml-org/llama.cppb9964Jul 11, 2026by github-actions[bot]same day as b9963

What changed since b9963

low risk

b9964 is a single-fix patch that stops argument handling from downloading spec models more than once.

Bug fixes

  • Argument parsing now avoids triggering duplicate downloads of speculative (spec) models when the same model is referenced more than once.

Upgrade notes

This is a drop-in patch over b9963 with no config or API changes; users running speculative decoding with repeated spec-model arguments should upgrade to avoid redundant downloads. No migration steps are required.

Analyzed Oct 24, 2026

Upgrading?

Release notes summary

Fixes an issue where duplicate downloads of spec models could occur in arguments.

Key Highlights

  • Prevents duplicate spec model downloads

New Features

  • Argument handling fix

Release Notes (excerpt)

<details open> arg: prevent duplicate spec model downloads (#25527) </details> **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) **Windows:**…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories