What is Eleven v4?
Eleven v4 is the first ElevenLabs speech model on a new architecture. It reads a script in context, follows square-bracket audio tags such as [whispers], [long pause] or [door slams], and can voice several speakers and sound effects in one generation, in 90+ languages and up to 10,000 characters per request. A faster sibling, Eleven v4 Turbo, answers live voice agents in about 100 ms.
It replaces Eleven v3, which ElevenLabs now calls its previous generation. Professional Voice Clones work again, but clones made before v4 need retraining, and v4 offers only Stability and Similarity settings, with no Style or Speed sliders and no SSML.
What changed from Eleven v3?
A new architecture with steadier voices, longer requests, more languages and Professional Voice Clones back, at the same price per character.
- Request length
- Up to 10,000 characters per request, against 5,000 on v3.
- Languages
- 90+ languages, against 70+ on v3.
- Consistency
- ElevenLabs says the same voice now holds across regenerations, dialogue and narration.
- Voice clones
- Professional Voice Clones work again, but clones made before v4 need retraining and may sound different.
- Controls
- Only Stability and Similarity settings; no Style or Speed sliders and no SSML.
- Price
- Both cost $0.08 per 1,000 characters on the API.
Who is Eleven v4 best for?
Eleven v4 is best for audiobooks, games, dubbing and voiceovers that need acted, consistent delivery in many languages, and, through v4 Turbo, for live voice agents.
Audiobooks and long narration
Context stitching keeps pacing steady across a long script, and ElevenLabs says the same voice holds up however many times you regenerate a line.
Games and character dialogue
Tags for emotion, reactions and sound effects, plus multi-speaker dialogue, let one generation act out a whole scene.
Dubbing and multilingual voiceovers
A voice recorded in one language speaks any of the 90+ others with a native accent while keeping its identity.
Live voice agents
Eleven v4 Turbo streams speech with about 150 ms to first sound at $0.04 per 1,000 characters, and works in ElevenAgents.
What's included with Eleven v4?
Audio tags for emotion, pauses and sound effects, multi-speaker dialogue, context stitching for long scripts, instant and professional voice cloning, pronunciation control and studio or phone audio formats.
- Audio tags
- Tags in square brackets set emotion, pauses, reactions and sound effects inside the script, and v4 follows them more reliably than v3.
- Multi-speaker dialogue
- The Text to Dialogue API voices a scene with several speakers in one generation.
- Context stitching
- Chains generations for long scripts so pacing and delivery stay consistent from start to finish.
- Voice cloning
- Instant Voice Clones from as little as 10 seconds of audio, and Professional Voice Clones, each needing the voice owner's verified consent.
- Pronunciation control
- Pronunciation dictionaries and IPA fix how names, acronyms and technical terms are said.
- Output formats
- MP3, WAV and PCM for studio work, and µ-law for phone systems.
How much does Eleven v4 cost?
Eleven v4 costs $0.08 per 1,000 characters on the ElevenLabs API, or $0.04 for v4 Turbo; the free plan includes 10,000 credits a month, and paid plans start at $6 a month.
- 10,000 credits a month, about 10 minutes of speech
- Personal use only
- No credit card required
- Eleven v4 Turbo: $0.04 per 1,000 characters
- Up to 10,000 characters per request
- Streaming and non-streaming endpoints
- 30,000 credits a month, about 30 minutes
- Commercial license
- Instant voice cloning
- 121,000 credits a month, about 121 minutes
- Professional voice cloning
Text to speech uses one credit per character, and credits are shared across all ElevenLabs products. Pro is $99 a month (600,000 credits), Scale $299 (1.8M credits, 3 seats) and Business $990 (6M credits, 10 seats); Enterprise is priced on request. Yearly billing gives two months free, so Starter works out to $5 a month, Creator $18.33 and Pro $82.50. On the API, each plan's included characters follow the per-character rate. Startups can apply for a grant of 33 million characters over 12 months. Prices are in US dollars and exclude tax. Full pricing on elevenlabs.io. We checked these prices on .
How do readers rate Eleven v4?
No one has rated Eleven v4 yet. Be the first.
Sign in to rate. One rating per account.
Eleven v4 pros and cons
Early listeners rate it the most convincing ElevenLabs voice yet, with twice v3's request length, but existing clones need retraining, it has only two voice settings and it costs more per minute than Google's option.
Pros
- Early listeners on Reddit call it the most convincing text-to-speech they have heard, and some say they cannot tell it from a real voice.
- Users who found their v3 voices flat or behind open models say v4 gave them the range they wanted and stays consistent from take to take.
- Takes twice as much text per request as v3, covers 90+ languages and supports Professional Voice Clones again, at the same $0.08 per 1,000 characters.
- One model family covers produced audio and live agents, with v4 Turbo at about 100 ms and half the price.
Cons
- Cloned voices can come out different: early users say their clones did not sound like the original or sounded better on v2 or v3, and ElevenLabs says older clones must be retrained for v4.
- Only two voice settings, Stability and Similarity: there are no Style or Speed sliders and no SSML, so pauses and pacing come from tags and punctuation.
- A voice speaking another language always drops its source accent; ElevenLabs says an on and off switch is still research, and one early tester found dialects hard to keep consistent.
- It costs about three times as much per minute of audio as Gemini 3.8 Flash TTS at Google's paid rate.
What are the best Eleven v4 alternatives?
Four tools on Recaply do Eleven v4's job. Qwen3-TTS is open source, so you can run it on your own server.
Gemini 3.8 Flash TTSFreeChoose it when you want a free API tier or a lower price per minute: Google's paid rate works out to about $0.027 a minute of audio, roughly a third of Eleven v4's, in 130 languages.
Sonic 3.6FreeChoose it when you want more minutes on a small plan: Cartesia's $5 Pro plan includes about 133 minutes of speech a month, against about 30 on ElevenLabs' $6 Starter.
Voxtral TTSFreeChoose it when price per character matters and your languages are covered: Mistral charges $0.016 per 1,000 characters, a fifth of Eleven v4's rate, with voice cloning in 9 languages.
Qwen3-TTSOpen SourceChoose it when you want Apache-licensed speech models for voice cloning and preset voices in 10 languages.
Questions about Eleven v4
What is the difference between Eleven v4 and Eleven v4 Turbo?
Eleven v4 is tuned for recorded work such as audiobooks, games and voiceovers, and costs $0.08 per 1,000 characters on the API. Eleven v4 Turbo is the low-latency version for voice agents, at about 100 ms median inference latency and $0.04 per 1,000 characters.
Who is Eleven v4 for?
Eleven v4 is best for audiobooks, games, dubbing and voiceovers that need acted, consistent delivery in many languages, and, through v4 Turbo, for live voice agents.
Is Eleven v4 free?
Yes. Eleven v4 has a free plan: 10,000 credits a month, about 10 minutes of speech.
How much does Eleven v4 cost?
Eleven v4's paid plans are API, pay as you go at $0.08 per 1,000 characters, Starter at $6 per month and Creator at $22 per month.
Is there an open source alternative to Eleven v4?
Yes. Qwen3-TTS is an open source tool that does the same job, so you can run your own.
What are the best Eleven v4 alternatives?
The closest alternatives on Recaply are Gemini 3.8 Flash TTS, Sonic 3.6, Voxtral TTS and Qwen3-TTS.
Where can I try Eleven v4?
On its official website, elevenlabs.io.
Do I need to retrain my voice clones for Eleven v4?
Yes. ElevenLabs says Instant and Professional Voice Clones made before v4 need retraining to work well. For a Professional Voice Clone, open My Voices, hover over the voice and click the plus button next to Eleven v4.
Can I use Eleven v4 audio commercially?
Only on a paid plan. ElevenLabs' Free plan is for personal use; the Starter plan at $6 a month adds a commercial license and instant voice cloning, and every higher plan keeps it.
Does Eleven v4 support SSML?
No. ElevenLabs disabled SSML tags such as break in Eleven v4, so you write pauses as audio tags like [pause] or [long pause]. Pronunciation dictionaries still work for names and technical terms.
Our verdict
Eleven v4 is the ElevenLabs model to use now for acted speech in audiobooks, games and dubbing.
It takes twice v3's text per request, covers 90+ languages, supports Professional Voice Clones again, and costs the same $0.08 per 1,000 characters on the API.
- Skip it for
- Cheap high-volume speech, where Gemini 3.8 Flash TTS costs about a third as much per minute.
- Check first
- Retrain and test your existing voice clones on v4 before you switch, since they can sound different.
Eleven v3
Eleven v4 Turbo
ElevenLabs Voice Design V3
Eleven Realtime
Scribe v2
Scribe v2 Medical