Skip to main content
← Back to news

llama.cpp v0.8.0: the LFM3 family, OpenAI realtime and TTS in the server

Tools & releases
Source:GitHub
The local inference engine adds the LFM3 family (1.7B text, 2B vision, 2B audio, 2B omni with ASR, TTS and voice chat), an OpenAI-realtime-protocol /v1/realtime endpoint and a dedicated POST /tts, plus new architectures and a hybrid cache for host-resident MoE experts.

Get the next local-AI briefing

Once a week, the top selection from iaLocale: articles, news and resources on local AI, and more.

After signing up you receive a confirmation email (double opt-in). Without confirmation we send nothing. No spam, we promise.