llama.cpp reaches v1.0 as a stable base for local AI
The ggml-org project has released llama.cpp and ggml version 1.0.0. The project says it has grown over 3.5 years and that future major-version releases will mark API and ABI incompatibilities. The project says llama.cpp supports more than 150 architecture variants and all mainstream speculative decoding methods, and it commits to continued support for all model architectures.