Eleven v3 Conversational
Eleven v3 Conversational brings the v3 performance approach to interactive speech. It was named in the February 2026 agent upgrade, alongside improvements to turn-taking.
Delivery in an agent
The February API update added the model identifier and suggested audio-tag settings for agents. Suggested tags guide delivery; they are not a substitute for the agent’s instructions or tools.ElevenLabs: v3 Conversational API support
Practical guidance. This model supplies an agent’s generated voice. Eleven Realtime configurations connect several components.ElevenLabs model catalog and deprecations
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | API, Provider product |
| Model identifiers | eleven_v3_conversational |
| Server latency | About 165 ms median, for warm voices; excludes network and application timeElevenLabs latency measurement method |
| Language coverage | More than 70 languagesElevenLabs API introduction and model specifications |
Related models
- Eleven v4 Turbo
ElevenLabs · Eleven speech
Low-latency v4 speech model for expressive interactive voices.
Launch date API · Provider product - Eleven Flash v2.5
ElevenLabs · Eleven Flash
Multilingual speech model designed for short response delays.
Launch date API · Provider product - Eleven v3
ElevenLabs · Eleven speech
Expressive speech model with audio tags and multi-speaker performance.
Launch date API · Provider product