Last refreshed 2026-10-20. Next refresh: weekly.
Why use Eleven Realtime v1 Mini on ElevenLabs API?
ElevenLabs API offers Eleven Realtime v1 Mini with pay-as-you-go pricing at $0.20/1M input tokens. ElevenLabs' API provides hosted text-to-speech, speech-to-speech, dubbing, and voice models for creative production and realtime conversational applications.
Setup recipe
Docs fallbackUse the provider REST API or SDKCreate a provider API keymodel: eleven_realtime_v1_minieleven_realtime_v1_miniRequest example
eleven_realtime_v1_mini.Gotchas
- Use provider model ID "eleven_realtime_v1_mini", not the LLMReference slug "eleven-realtime-v1-mini".
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.20 |
| Output tokens | $6.00 |
Capabilities
About Eleven Realtime v1 Mini
Eleven Realtime v1 Mini is ElevenLabs' default speech-to-speech voice agent model, released October 20, 2026. It packages speech recognition, voice activity detection and turn-taking, a language model and text-to-speech as one system tuned end to end (a cascaded architecture; ElevenLabs does not disclose the component models). It is optimized for low latency, cost and natural conversation, including interruption handling, supports tool calling and 90+ languages, and ElevenLabs reports 620 ms time to first audio. It runs in ElevenAgents and through an OpenAI Realtime-compatible WebSocket API.