llama.cpp v1.0.0
ggml-org released llama.cpp v1.0.0 and ggml v1.0.0, the first major versions of the local LLM inference library and its tensor library, with semantic versioning from now on.
Both projects move to 1.0.0 after about 3.5 years of development. API and ABI incompatibilities will now be marked with major-version bumps, and the public C headers (libllama and libggml) define what counts as a breaking change. Minor versions cover backward-compatible features, backend updates and model support, and patch versions cover bug fixes.
- Date
- Friday 13 November 2026
- Lab
- ggml.ai (llama.cpp)
- Kind
- infra
- Access
- open weights
The software is open source. The access value 'open weights' is the closest allowed option and does not mean model weights were released.
Sources
This record was checked against its sources on 13 November 2026. How we check