llama.cpp v1.0.0
llama.cpp v1.0.0 released with ggml 1.0.0 and server, CUDA, Vulkan, and Hexagon updates.
The Deploy Log | Products | Free, no signup
llama.cpp changelog
5 versions of llama.cpp from llama.cpp in The Deploy Log. From to . On the median a new version every 11 days, publisher's dates only; the longest wait was 20 days. The newest is v1.0.0. 2 versions in the last 30 days. Each links the publisher's own page.
| Version | Publisher's date | Days after the one before |
|---|---|---|
| v1.0.0 | 9 | |
| v0.9.0 | 20 | |
| v0.7.0 | 10 | |
| v0.6.0 | 12 | |
| v0.5.0 |
llama.cpp v0.9.0
Adds OCP FP8 E4M3 quantization, flash_attn_ext_rows, four new architectures, Chatterbox TTS, and Gemma 4 E2B/E4B prefill optimizations.
llama.cpp v0.7.0
llama.cpp v0.7.0 adds LiliCorr speculative decoding, MTP for GLM5-Next, and a GPU cache for MoE experts with up to ~2.2x decode speedup.
llama.cpp v0.6.0
llama.cpp v0.6.0 adds GLM-5.3-Flash 320B, Clef, Qwen4Exp MTP, and a new /v1/systemone server API.
llama.cpp v0.5.0
llama.cpp v0.5.0 adds HRM-Text support, MiMo-V2.6 and HunyuanOCR conversion, and multi-address HTTP binding.
One line every time AI actually ships, around the clock, free on Threads and Bluesky. Every one is in The Deploy Log with its source.