Generate
Back to Models
Models/elevenlabs/modelPass-through

ElevenLabs v3

ElevenLabs v3 reads from the writing: emotion, emphasis, and pacing follow the punctuation and phrasing rather than a flat delivery. On MachGen one checkpoint serves two surfaces - speech in a single voice, up to 3,000 characters per request, and dialogue placing up to 10 distinct voices in one acoustic space, up to 2,000 characters. Voices come from the account library, with stability, similarity, style, and speed adjustable per request.

Current MachGen support
Text to SpeechText to DialogueUp to 10 voices3,000 characters per requestSeed control
Starting at $0.0001 / char
All supported sizes$0.0001/char
Live public pricebookSee full pricebook
Overview

ElevenLabs v3

ElevenLabs v3 reads from the writing: emotion, emphasis, and pacing follow the punctuation and phrasing rather than a flat delivery. On MachGen one checkpoint serves two surfaces - speech in a single voice, up to 3,000 characters per request, and dialogue placing up to 10 distinct voices in one acoustic space, up to 2,000 characters. Voices come from the account library, with stability, similarity, style, and speed adjustable per request.

ElevenLabs v3 API on MachGen. Compare supported generation modes, examples, and current pricing from the live pricebook.

Model guide

Why choose ElevenLabs v3

Key strengths

  • Expressive delivery: emotion, emphasis, and pacing follow the writing rather than a flat read.
  • Multi-speaker dialogue in one pass, so turns share an acoustic space instead of being stitched together.
  • A large voice library plus consistent identity across separate generations of the same voice.

Good fit for

  • Narration, voiceover, explainer, and audiobook passages.
  • Character dialogue for games, animation, and video scenes.
  • Localized reads of a script that must keep one performance style.
Supported on MachGen

Inputs and output settings

This table reflects the model routes and controls currently exposed by MachGen.

ModeRequired inputAvailable settings
Text to SpeechText prompt and a voice from the account's libraryVoice selection from the account library
Seed control
Prompt limit: 3,000 characters
Text to DialogueOrdered speaker turns, each pairing a voice with its line (up to 10 distinct voices)Voice selection from the account library
Seed control
Prompt limit: 2,000 characters
Workflow

Using ElevenLabs v3

  1. Choose a supported generation mode in Playground.
  2. Add the required prompt and source or reference assets.
  3. Select output settings from the controls shown for that mode.
  4. Generate, then inspect the returned asset before reusing the settings through the API.
Best results

How to get better ElevenLabs v3 results

  • Write punctuation the way it should be performed - it drives the pauses and the phrasing.
  • Use dialogue rather than several separate speech generations when speakers respond to each other.
  • Lower stability for emotional range; raise it when a long piece must stay uniform.
  • Keep a single request under the character cap and split long scripts at natural scene breaks.
Limits

Limits and availability

  • MachGen admits only the modes and settings listed above for this catalog model.
  • Source assets must finish uploading before a request can be submitted.
  • Generated results can vary between requests, including when the same prompt and settings are reused.
  • ElevenLabs v3 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.
Learn more

Learn more

For model background, technical details, and architecture, see the official ElevenLabs text to speech guide.

Try this model in the MachGen Playground.

Official documentation may describe capabilities or parameters that are not currently exposed by MachGen. Use the Inputs and output settings section above as the source of truth for this page.

Pricing

ElevenLabs v3 pricing

See Pricing for currently published MachGen rates. When available, the Playground estimate reflects the selected task and output settings.

On MachGen

ElevenLabs v3 on MachGen

Deployment: Pass-through. ElevenLabs v3 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

The Playground and API sections on this page list the inputs and output settings exposed for this model.