← All models

ElevenLabs Eleven v4

Audio Available Full comparison ↗

elevenlabs/eleven-v4 · by ElevenLabs · neural-tts

Pricing — 1 offering(s)

ElevenLabs

Text characters

  • $22.00 / 1M characters Current 2026-09-28 → 2026-10-12
    Launch promotion, time-limited: elevenlabs.io/pricing/api shows the "v4" card at "72% off until Oct 12"…

    Launch promotion, time-limited: elevenlabs.io/pricing/api shows the "v4" card at "72% off until Oct 12": $0.022 per 1K characters ($22/1M), with the $0.08 list price struck through. Recorded with its end date so the promo is never read as the permanent price (SCO-670). effective_from is the launch date (elevenlabs.io/blog/eleven-v4, 2026-09-28).

  • $80.00 / 1M characters Future 2026-10-13
    List price: the struck-through $0.08 per 1K characters ($80/1M) on the "v4" card during the launch promotion…

    List price: the struck-through $0.08 per 1K characters ($80/1M) on the "v4" card during the launch promotion, which ends 2026-10-12 (SCO-670). Re-verify once the promo ends.

Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.

See how ElevenLabs Eleven v4 fits into a cost-aware routing setup

See how →

Capability profile

How ElevenLabs Eleven v4 rates across core capability dimensions, with the task-level evidence behind each rating.

voice naturalness Strong

Vendor claims only so far: "Ranked #1 by Artificial Analysis" and preferred by ~75% of listeners in blind head-to-head tests (launch post). Not yet independently checked here; v3 was #12 of 98 on the Artificial Analysis TTS Arena.

language support Strong

90+ languages per ElevenLabs' launch post (v3: 70+).

cloning support Strong

Full ElevenLabs voice cloning; the launch post says Instant Voice Clones now need about 10 seconds of audio.

streaming latency Strong

~100ms median inference latency per the launch post (vendor-reported).

Ratings are estimated — limited independent data is available for this model.

Operator guidance

Choose Eleven v4 over v3 for new work: same $80/1M list price (a launch promotion runs to 2026-10-12), more emotive, and 90+ languages. Choose v4 Turbo when latency matters more than the last bit of expressiveness.

Use cases

Limitations

Citations