Eleven v4 Turbo
elevenlabs · Released Sep 2026
Conditional. Eleven v4 Turbo (id eleven_v4_turbo) is ElevenLabs' real-time speech model, stating a median inference latency of about 100 ms, excluding application and network latency, across 90+ languages via a Text to Dialogue WebSocket. The deciding caveat is that the latency figure is vendor-stated and untested by us, and $0.011 per 1K characters is a promo ending Oct 12 against a $0.04 list rate. Suits builders of voice agents and interactive characters.
Is it right for you?
Good for
- Real-time voice: the models page states a median inference latency of about 100 ms, excluding application and network latency
- Support agents, AI assistants and interactive characters, the use cases the models page names
- Streaming over the Text to Dialogue WebSocket, the endpoint the changelog lists for agents and interactive applications
- Multilingual output: 90+ languages, the same roster the models page lists for eleven_v4
Not good for
- Long single-shot generations: the models page we read states no character limit for eleven_v4_turbo, so confirm before relying on long inputs (eleven_v4 lists 10,000 characters)
- Budgets built on the launch promo rate: the promo has ended and the pricing page shows $0.04 per 1K characters as the price
- Latency budgets that must include network and application time: the 100 ms figure excludes both
How it performs by task
Real-time voice agents
ElevenLabs states about 100 ms median inference latency and names support agents and AI assistants as the use; vendor-stated, untested by us
Expressive speech
ElevenLabs calls it its most emotive real-time speech synthesis model; the pages we read give no independent score
Voice cloning
The models page lists exceptional voice cloning; we have not tested it
Pricing
Input
$0.04 / 1K characters
Output
Billed per character (single rate)
Context
List $0.04 / 1K chars; median latency ~100 ms
No verdict changes yet
The clock starts day one. Changes land here as our verdict evolves.
Sources
Verification log
- Pricing· Updated
Automated agent
List price $0.04 per 1K chars confirmed, no promo text on the page. Promo ended Oct 12. Refreshed stale pricing_context and the promo avoid_for row. Standing verdict_summary still describes the promo and needs a staged same-rating rewrite at publish.
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent