b10344

ggml-org/llama.cppb10344Aug 10, 2026by github-actions[bot]same day as b10343

What changed since b10343

low risk

Adds Multi-Token Prediction (MTP) support for Nemotron Nano models, with no other changes between builds.

New features

  • Nemotron Nano models can now use Multi-Token Prediction (MTP) for speculative-style generation.
  • New `mtp_flags` option exposed for Nemotron models to control MTP behavior.

Upgrade notes

This is a single-commit delta over b10343, so only Nemotron users gain anything new; all other behavior is unchanged. If you serve Nemotron Nano models, review the new `mtp_flags` setting to enable and tune MTP. No config migration is required for other model families.

Analyzed Nov 1, 2026

Upgrading?

Release notes summary

Adds Multi-Token Prediction (MTP) support specifically for the Nemotron model family, enhancing generation capabilities for these models.

Key Highlights

  • Added MTP support for Nemotron Nano models.
  • Added `mtp_flags` for Nemotron models.

Release Notes (excerpt)

<details open> model: add MTP support for Nemotron model (#26725) * model: add MTP support for Nemotron Nano model * model: add mtp_flags for nemotron model * address review comments </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)…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories