Ten

TEN is an open-source framework for building real-time multimodal conversational AI agents that can see, hear, and speak with users. It features a modular architecture that seamlessly integrates large language models with speech recognition, text-to-speech, vision processing, and real-time communications capabilities. Developers can create agents with natural voice interactions, visual understanding, and even animated avatars while easily swapping AI components through plug-and-play extensions without code changes. TEN distinguishes itself with its visual graph-based configuration system, support for cutting-edge real-time AI services like Gemini 2.0 Live and OpenAI Realtime, and compatibility with platforms like Dify and Coze. Organizations seeking low-latency conversational agents with multimodal capabilities will appreciate TEN's comprehensive AI stack that combines the flexibility of open-source development with production-grade performance for applications requiring natural human-AI interaction.