ElevenLabs launches two real-time speech-to-speech models for AI conversations

ElevenLabs released two new speech-to-speech models, Eleven Realtime v1 Max and v1 Mini, designed for real-time AI conversations. The v1 Max model scores 98.8 percent on the Big Bench Audio benchmark with a 0.57-second response time, positioning it as more accurate and faster than other models on the Artificial Analysis leaderboard. The smaller v1 Mini is built for natural, lower-cost conversations and can detect when a user has finished speaking. Both models are available through the ElevenAgents platform and an OpenAI-compatible Realtime API, with pricing based on per-minute costs plus token usage.