









Hathora Models provides a platform for building voice agents on open-source or closed models with zero DevOps. It offers low-latency ASR (Automatic Speech Recognition), TTS (Text-to-Speech), and LLM (Large Language Model) models that run in 14 regions for ultra-low latency. Users can start instantly on shared endpoints and upgrade to dedicated infrastructure for privacy, compliance, or VPC requirements. The platform allows users to explore, test, and deploy production-ready models, bring their own models or custom containers, and utilize a "Chain tool" for interactive voice AI pipelines.
To use Hathora Models, first, discover models by browsing open-source STT, TTS, and LLM models hand-picked for voice AI use cases. Next, test models instantly in interactive sandboxes or swap them seamlessly in the Chain tool. Finally, deploy fast using provided documentation for Pipecat, LiveKit, and direct API access.





Social Listening