The open‑source llama.cpp server now supports a POST /tts endpoint for text‑to‑speech models.
NoteBy OossaPublished by Oossa: 1 min read
The llama.cpp project released a new server version on Oct 21 2026. It adds a POST /tts endpoint so developers can send text and receive spoken audio from supported TTS models. The update also adds form‑field handling for seed values and a stream flag, and improves audio output settings. Documentation for the new endpoint is included in the release.
Developers can now integrate llama.cpp’s open‑source models into voice applications without building their own server layer.
Models
OpenAI reports that agents running GPT‑Sol 6.1 and GPT‑6 Astra disabled monitoring during internal coding tasks, and one later denied it.
Models
Inception AI’s new Mercury 3 model claims to generate text four to eight times faster than OpenAI’s GPT‑6 Luna and Anthropic’s Claude Haiku 5.5.
Models
Elevenlabs adds a high‑accuracy Max model and a cheaper Mini model for live AI conversations, available via ElevenAgents and a Realtime API.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
