Eleven Realtime v1 Max

Use it to build voice agents for customer support, lead qualification, and multi-step workflows that require reasoning or heavy tool use.

Availability

Launching today; available in ElevenAgents and through the Realtime API

Pricing model

Not established

Deployment

ElevenAgents production deployment; Realtime API over WebSocket

Privacy

ElevenAgents can apply guardrails and PII redaction to a Realtime agent

Licence

Not established

• Eleven Realtime v1 Max is available in ElevenAgents and through the Realtime API.
• Both Eleven Realtime models support more than 90 languages and can use any of the 17,000+ voices in the ElevenLabs library. ^[1]

  Both support more than 90 languages and can use any of the 17,000+ voices in the ElevenLabs library:

• v1 Max answered 98.8% of Big Bench Audio questions correctly. ^[1]

  Realtime v1 Max answered 98.8% questions correctly, and v1 Mini answered 75.9% correctly.

• v1 Max starts speaking in 0.57 seconds on average in the reported Big Bench Audio test. ^[1]

  V1 Max starts speaking in 0.57 seconds on average, and v1 Mini in 0.62 seconds.

• v1 Max scored 94.1% on Full-Duplex-Bench. ^[1]

  On Full-Duplex-Bench, which tests pauses, turn-taking, interruptions, and backchannels, v1 Mini scores 95.8% and v1 Max 94.1%.

• v1 Max completed 87.1% of tool-use tasks. ^[1]

  On tool-use tasks, v1 Max completed 87.1% and v1 Mini 60.3%, which is why we recommend Max for use cases with heavy tool calling.