Skip to main content

audio model comparison

Eleven Multilingual v2 vs Eleven v4

Eleven Multilingual v2 and Eleven v4 are both current audio models, from ElevenLabs. Both cost $0.00008 per character. On Artificial Analysis Speech Arena, Eleven Multilingual v2 scores 1096 Elo and Eleven v4 1326 Elo (Artificial Analysis, measured October 13, 2026).

Data verified October 13, 2026 · Dataset 2026-10-13

Share this

Side by side

Standard-tier prices in effect today. The better value in each row is highlighted.

FactEleven Multilingual v2Eleven v4
Price per character$0.00008$0.00008
Artificial Analysis Speech Arena1096 Elo (#8 of 11)1326 Elo (#2 of 11) (better)
ProviderElevenLabsElevenLabs
Release dateNot reportedSeptember 28, 2026
AccessAPI onlyAPI only
Context windowNot reportedNot reported
Max outputNot reportedNot reported
Inputtexttext
Outputaudioaudio

Scores:Artificial Analysis · 2026-10-13

Open both in the interactive comparison

When to choose each

Eleven Multilingual v2

Best for

  • Long-form multilingual narration

Watch out for

  • ElevenLabs now calls it a previous-generation model, priced the same as Eleven v4
Full Eleven Multilingual v2 profile

Eleven v4

Best for

  • Expressive narration and dialogue
  • Multilingual voice content

Watch out for

  • ElevenLabs' blind-test preference claims are its own; the independent rating is recorded below
  • For low latency ElevenLabs points to Eleven v4 Turbo
Full Eleven v4 profile

Eleven Multilingual v2 vs Eleven v4 FAQ

Which is cheaper, Eleven Multilingual v2 or Eleven v4?

Both cost $0.00008 per character.

Which scores higher on independent benchmarks, Eleven Multilingual v2 or Eleven v4?

Eleven v4 leads on Artificial Analysis Speech Arena, 1326 Elo to 1096 Elo. Scores are independent evaluations, not either vendor's launch figures.

Production evaluation

Choosing between Eleven Multilingual v2 and Eleven v4?

Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.

  • A head-to-head evaluation on your real workload
  • Routing between models by cost and difficulty