b10261

ggml-org/llama.cppb10261Aug 4, 2026by github-actions[bot]same day as b10259

What changed since b10259

low risk

b10261 adds Plamo2 byte-token vocabulary validation on top of b10259's tensor-reshape loading support.

New features

  • Plamo2 vocabularies now undergo byte-token validation during loading.

Improvements

  • Vocabulary integrity checks for Plamo2 models catch malformed byte tokens earlier.
  • Model loading remains more flexible thanks to the tensor reshape support introduced in the prior build.

Upgrade notes

This is a two-build delta with only one documented change, so treat any intermediate commits as unverified. Plamo2 models with malformed byte tokens that previously loaded silently may now fail validation at load time. No config or API changes are required; just rebuild or pull the new binaries.

Analyzed Nov 1, 2026

Upgrading?

Release notes summary

Adds validation for byte tokens in the Plamo2 vocabulary.

Key Highlights

  • Validation of Plamo2 byte tokens
  • Improved vocabulary integrity

Release Notes (excerpt)

<details open> vocab : validate plamo2 byte tokens (#26511) * validate plamo2 byte tokens * --typo </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) - Ubuntu x64 (SYCL FP32)…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories