b10195

ggml-org/llama.cppb10195Jul 30, 2026by github-actions[bot]same day as b10194

What changed since b10194

low risk

b10195 is a test-only build change: removes get-model.cpp to cut redundant builds and fixes quant type selection in tests.

Improvements

  • Test builds no longer recompile get-model.cpp repeatedly, reducing CI/build time.
  • Quantization type selection in test scenarios is now handled correctly.

Bug fixes

  • Corrected quant type selection logic used by tests, which previously could pick the wrong type.

Upgrade notes

No runtime, API, or model-format changes are involved; the delta is confined to the test suite and build graph. Users can upgrade from b10194 without any config or migration work. The CUDA transpose-free gemmv optimization landed in b10194 and remains unchanged here.

Analyzed Nov 1, 2026

Upgrading?

Release notes summary

This release focuses on optimizing the build process for tests by removing redundant files and fixing quantization type selection logic. It removes the `get-model.cpp` file to prevent redundant builds and ensures correct quant type selection in test scenarios.

Key Highlights

  • Removed `get-model.cpp` to avoid redundant builds during tests
  • Fixed quant type selection in test cases

New Features

  • Test infrastructure improvements

Release Notes (excerpt)

<details open> tests : avoid building get-model.cpp many times (#26317) * tests : remove get-model.cpp * tests : fix quant type selection </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