ElevenLabs releases Eleven Realtime speech-to-speech model
Understand the story
On October 20, 2026, ElevenLabs released two speech-to-speech models for real-time AI conversation, Eleven Realtime. The v1 Max scored 98.8% on the Big Bench Audio benchmark with a response time of 0.57 seconds; the company said that on the Artificial Analysis leaderboard, no model is currently both faster and more accurate. The other, v1 Mini, targets low-cost natural conversation scenarios and can distinguish whether the other party has truly finished speaking or merely made backchannel sounds like "mm-hmm." This is the first report on the event; no follow-up developments yet.
AI overview of reporting · Updated 2 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- The DecoderElevenlabs releases two new speech-to-speech models designed for real-time AI conversations
On October 20, ElevenLabs released two speech-to-speech models under the Eleven Realtime name: v1 Max scored 98.8% on the Big Bench Audio benchmark with a 0.57-second response time, and the company says no model on the Artificial Analysis leaderboard is currently both faster and more accurate; v1 Mini targets low-cost natural conversation and can distinguish when the other party has finished speaking from a mere mm-hmm.
Story heat trend
Current heat9·Comparable peak10（10/21 04:00）·Comparable change over 24 hours –
The trend compares the same continuously observed sources, which may cover fewer sources than the current heat score. Hover or select the chart to inspect each hour; use the left and right arrow keys to navigate.