DNA News

Act on this, mostly signal

ElevenLabs ships two real-time speech-to-speech models

What to do

Test Eleven Realtime v1 Max or Mini in ElevenAgents or the OpenAI-compatible Realtime API.

Why

Max scores 98.8% on Big Bench Audio with a 0.57-second response time.

Read the source at the-decoder.com Discussion

What the research found

ElevenLabs released Eleven Realtime v1 Max and v1 Mini speech-to-speech models, available via ElevenAgents and an OpenAI-compatible Realtime API, with Max at 98.8 percent on Big Bench Audio and 0.57-second response time.

Claims and how well the evidence supports them

  1. Eleven Realtime v1 Max scores 98.8 percent on Big Bench Audio with 0.57-second response time 99%
  2. Both v1 Max and v1 Mini are available through ElevenAgents and an OpenAI-compatible Realtime API 98%
  3. Pricing is per minute plus token costs, with v1 Max costing more than v1 Mini 99%

Evidence

Actionability rated medium.