Eleven Realtime is ElevenLabs' new family of speech-to-speech models built for voice agents. Speech recognition, turn-taking, the language model, and the voice are all tuned end-to-end to work well together. v1 Mini is designed for natural conversation and reaches first audio in about 620ms, while v1 Max handles more complex tasks and comes in around 570ms. The models support over 90 languages and work inside ElevenAgents or through an OpenAI-compatible Realtime API. To switch, you only change the endpoint and model name.
Categories:
Artificial Intelligence
Launch Date:
October 27, 2026