b9789
ggml-org/llama.cppb9789Jun 25, 2026by github-actions[bot]same day as b9788
What changed since b9788
b9789 is a single-purpose patch that fixes quantization of MoE models when MTP is enabled.
Bug fixes
- Quantization no longer produces errors for Mixture of Experts models used with MTP.
Upgrade notes
This is a narrow quantization fix with no CLI, config, or API changes, so upgrading from b9788 is a drop-in binary swap. Users who quantize MoE models with MTP enabled should re-run quantization with b9789 to get correct output; the SYCL tensor-parallelism work from b9788 is unchanged.
Analyzed Oct 31, 2026
Upgrading?
Release notes summary
Fixes a quantization bug affecting Mixture of Experts (MoE) models when using MTP (Mixture of Token Pruning).
Key Highlights
- Fix quantizing MoE with MTP
- Resolves quantization errors in MoE models
New Features
- MoE quantization bug fix for MTP
Release Notes (excerpt)
<details open> quant : fix quantizing moe with mtp (#24986) </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 notesRelated Repositories
:electron: Build cross-platform desktop apps with JavaScript, HTML, and CSS
Godot Engine – Multi-platform 2D and 3D game engine
ClickHouse® is a real-time analytics database management system
Self-hosted game stream host for Moonlight.
DuckDB is a high-performance analytical database system