ggml-org announced llama.cpp and ggml v1.0.0 on November 13, 2026, with a versioning commitment for developers building on the libraries. Future incompatibilities in the programming interfaces applications use (APIs) or the binary interfaces used by compiled software (ABIs) will come with major-version bumps, the team says. Breaking changes aren't being banished. They're supposed to arrive wearing a bigger number. Source
llama.cpp is a portable C/C++ implementation for running multimodal large language models across hardware ranging from consumer machines to enterprise servers. Its underlying machine-learning engine, ggml, is a tensor library; the releases cover libllama and libggml, respectively. Source
The team says many applications already depend on its work, making stable interfaces increasingly important. Source
What the version numbers promise
The announcement sets out three rules for ggml-org projects using semantic versioning:
- Major: A breaking change to the public C API in
./include. Source
- Minor: Backward-compatible features, backend updates, model support or API additions. Source
- Patch: A bug fix with no API change. Source
The table specifies the public C API boundary, while the broader release commitment also covers ABI incompatibilities. Source
“As adoption grows, we will be stricter about making those boundaries visible to downstream consumers,” the team writes. Source
A major-version bump is a cue to check an integration before upgrading. Source
Comments