ElevenLabs Realtime (v1 Mini, v1 Max and Realtime Translation)
ElevenLabs' speech-to-speech models (v1 Mini, v1 Max) with ~570-620 ms to first audio, plus a 90+ language live translation API.
This entry is marked for review. Check the original sources before relying on its details.
Overview
ElevenLabs shipped two realtime speech-to-speech models, v1 Mini (conversation) and v1 Max (complex tasks), listed on October 20, 2026, with reported time-to-first-audio of 620 ms and 570 ms respectively. They run inside ElevenAgents or through an OpenAI-compatible Realtime API, so existing clients only change the endpoint and model name, and deploy to phone, web widget, WhatsApp and contact-center platforms. A companion Realtime Translation layer streams bidirectional speech-to-translated-speech over a single WebSocket across 90+ languages, keeping one consistent voice (library or cloned) and streaming captions. ElevenLabs lists SOC 2, HIPAA and GDPR support with EU data residency and zero-retention modes.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- v1 Mini: 620 ms time-to-first-audio; v1 Max: 570 ms (vendor figures)
- OpenAI-compatible Realtime API; works in ElevenAgents
- Realtime Translation over a single WebSocket, 90+ languages
- Consistent voice across languages, including cloned voices
- Streaming captions and transcripts
- SOC 2, HIPAA, GDPR; EU data residency and zero retention
Use cases
- Voice agents on phone lines and WhatsApp
- Live interpretation between two speakers in a call
- Drop-in replacement for an OpenAI Realtime endpoint
Pricing
unknown; Product Hunt listings show a free option but no prices.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 20 Oct 2026
- Record updated
- 28 Oct 2026
- Record last checked
- 28 Oct 2026