●Change8

Migrating to llama.cpp b9700

Version b9700 introduces 2 breaking changes. This guide details how to update your code.

Released: 6/18/2026

2
Breaking Changes
2
Migration Steps
4
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

GGML_SYCL_SUPPORT_LEVEL_ZEROGGML_SYCL_SUPPORT_LEVEL_ZERO_APIGGML_SYCL_ENABLE_LEVEL_ZEROGGML_SYCL_USE_LEVEL_ZERO_API

Breaking Changes

●Issue #1

The SYCL related macro GGML_SYCL_SUPPORT_LEVEL_ZERO has been renamed to GGML_SYCL_SUPPORT_LEVEL_ZERO_API. Users relying on this macro must update their code.

●Issue #2

The SYCL related macro GGML_SYCL_ENABLE_LEVEL_ZERO has been renamed to GGML_SYCL_USE_LEVEL_ZERO_API. Users relying on this macro must update their code.

Migration Steps

  1. 1
    Rename usages of GGML_SYCL_SUPPORT_LEVEL_ZERO to GGML_SYCL_SUPPORT_LEVEL_ZERO_API.
  2. 2
    Rename usages of GGML_SYCL_ENABLE_LEVEL_ZERO to GGML_SYCL_USE_LEVEL_ZERO_API.

Release Summary

This release primarily renames SYCL-related macros for clarity and includes minor fixes for code formatting and rebase errors. New binaries are provided across multiple platforms and accelerators.

Need More Details?

View the full release notes and all changes for llama.cpp b9700.

View Full Changelog