Eleven Flash v2.5

Ultra-low latency TTS (~75ms) for conversational apps; 32 languages; leaner per-character cost.

Eleven Flash v2.5 is the ultra-low latency speech model (~75ms) with 32 languages and a leaner per-character API price. Official guidance points conversational apps here instead of Turbo.

Model spelling matches the official docs. For quotas, languages, and the latest changelog, check the official ElevenLabs website.

Open All models