Latest models
- Eleven v4 and v4 Turbo (September 28, 2026) Current
Text-to-speech models for expressive narration (v4) and low-latency agents (v4 Turbo), supporting more than 90 languages. Source: ElevenLabs blog ↗ - ElevenLabs Music v2.5 (September 2026) Current
The new default music model in ElevenMusic. ElevenLabs says you own what you make on every plan, including Free, with lossless downloads on all plans. Source: ElevenLabs blog ↗
Checked October 7, 2026. See the full AI model tracker.
What we like
- Expressive text-to-speech in more than 90 languages (Eleven v4)
- ElevenMusic: ElevenLabs says you own what you make on every plan, including Free
- Tools for dubbing, agents and voice cloning
Watch out for
- Pronunciation of names and jargon may need tweaking
- Voice cloning requires consent and the right to use that voice
- Usage limits depend on your plan
ElevenLabs is an AI audio platform. You type or paste a script and pick a voice, and it reads it back naturally. It also makes music and offers dubbing and voice-agent tools.
What is current
- Eleven v4 and v4 Turbo (September 28, 2026): text-to-speech models for expressive narration and low-latency agents, supporting more than 90 languages. See our news summary.
- Music v2.5 (September 2026): the new default in ElevenMusic. ElevenLabs says lossless downloads are available on every plan and that downloads are blocked on tracks that reference other artists’ songs.
Sources are in the model tracker.
How prompting works
Write the script the way you want it spoken: short sentences, clear punctuation, and phonetic spellings for tricky names. Try a few voices and settings. For music, describe genre, instruments, mood and vocals.
A word on consent
Only clone or imitate a voice when you have the speaker’s permission and the right to use it. Check the platform’s terms and the law where you live.
Compare music tools in Suno vs ElevenLabs Music.
Contribute your creativity