Cloudflare's multimodal decision model, released 9 October 2026 under Apache 2.0: takes text, images, audio and video and returns a score for each allowed answer instead of free text. 30B mixture of experts with 3B active.
Eleven v4
New· 4ElevenLabs' text-to-speech model for expressive speech in 90+ languages, with inline emotion tags and long-form stitching
About Eleven v4
Scripts with several speakers, shifting moods and sound effects are what Eleven v4 is built to read. ElevenLabs announced this text-to-speech model on 28 September 2026, and the company says it rests on a new architecture that follows a script roughly the way a voice actor would, keeping track of who is speaking and what has just happened in the scene.
Ninety languages and directed delivery
According to ElevenLabs, v4 produces speech in more than 90 languages. The delivery can be steered in plain language or with inline tags that call for an emotion, an accent or a sound effect. The company claims v4 obeys those directions more reliably than its v3 model, and says pronunciation set through International Phonetic Alphabet phonemes is also more dependable.
Staying consistent across long scripts
Audiobooks and other long-form work are where pacing tends to drift. Eleven v4 uses context stitching to keep the rhythm steady from one section to the next, and regenerating a single line keeps the same voice instead of changing it. Professional Voice Clones, which v3 could not use, work with v4 again.
Where it is available
The model can be used in ElevenLabs' creative tools, on its ElevenAgents platform and through its API. It is a hosted model only: no weights are published, so it cannot be run on a company's own hardware. A lower-latency variant, Eleven v4 Turbo, was released at the same time, and the earlier Multilingual v2 remains part of the same lineup.
How to read the claims
ElevenLabs points to a third-party ranking and to its own blind listening tests as evidence of quality. Those are the company's statements, and buyers who care about a particular language or voice style would do well to audition it with their own scripts before committing.
TrustList summary
Eleven v4 is a text-to-speech model from ElevenLabs, announced on 28 September 2026 by its co-founders. The company says the model interprets a script as a voice actor would, considering the speaker and what has just happened, and it produces speech in more than 90 languages. It is hosted only, with no published weights.
Main features
- Directed delivery: emotion, accent and sound-effect direction, given in plain language or with inline tags.
- Pronunciation control: International Phonetic Alphabet phonemes, which the company says work more reliably than in v3.
- Long-form audio: context stitching keeps pacing steady across long scripts such as audiobooks, and regenerating a line keeps the same voice.
- Voice cloning: Professional Voice Clones, which v3 did not support, work again.
Who it serves
Teams producing narrated or spoken audio, and developers building voice agents.
Key facts
- Released: 28 September 2026
- Languages: more than 90
- Access: ElevenLabs' creative tools, the ElevenAgents platform and the API
- Companion model: Eleven v4 Turbo, a low-latency variant released at the same time
The company's listening tests and a third-party ranking it cites have not been checked independently.
Summary written by TrustList from the business's own published information, Nov 2026.
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| No benchmark scores yet. Be the first to add one. | ||
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
Trust Score
An earned signal from verification, reviews, awards, transparency and engagement - the vendor can't buy it.
Updated 10/3/2026
What the business tells TrustList
- Founded
- 2026 (0 years) evidence sent
Request a demo or quote from Eleven v4
Reviews
Write the first review of Eleven v4
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about Eleven v4.
Eleven v4 alternatives
Similar listings buyers compare against this one.
Smaller, latency-optimised Cloudflare decision model that returns probability-scored classifications of text or images
Perplexity's open-weight decision model that picks among given options for routing and checks in AI systems
Cantina Security's first open-weight security model, a post-trained GLM-5.3-Flash for vulnerability research and exploit checks
Black Forest Labs' image generation and editing model with bounding-box layouts, multiple reference images and 4K output
Microsoft AI's real-time speech-to-text model for voice agents, transcribing 60 languages with continuous language detection
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.