ElevenLabs launches Eleven Realtime speech-to-speech models for voice agents

ElevenLabs launched Eleven Realtime, speech-to-speech models that combine speech recognition, turn-taking, a language model and voice in one agent.

v1 Mini is the lower-cost option for assistants, tutors and interactive characters. v1 Max reasons before responding and is aimed at customer support and multi-step tasks with tool calls. Both support 90+ languages and voices from the ElevenLabs library.

ElevenLabs says v1 Max starts speaking in 0.57 seconds on average and v1 Mini in 0.62 seconds. Both are available in ElevenAgents and through a Realtime API compatible with OpenAI’s Realtime API.