audio model comparison
Eleven v4 vs Sonic 3.6
Eleven v4 and Sonic 3.6 are both current audio models, from ElevenLabs and Cartesia. Sonic 3.6 is 19% cheaper per character than Eleven v4, $0.000065 vs $0.00008. On Artificial Analysis Speech Arena, Eleven v4 scores 1326 Elo and Sonic 3.6 1286 Elo (Artificial Analysis, measured October 13, 2026).
Data verified October 13, 2026 · Dataset 2026-10-13
Side by side
Standard-tier prices in effect today. The better value in each row is highlighted.
| Fact | Eleven v4 | Sonic 3.6 |
|---|---|---|
| Price per character | $0.00008 | $0.000065 (better) |
| Artificial Analysis Speech Arena | 1326 Elo (#2 of 11) (better) | 1286 Elo (#3 of 11) |
| Provider | ElevenLabs | Cartesia |
| Release date | September 28, 2026 | August 27, 2026 |
| Access | API only | API only |
| Context window | Not reported | Not reported |
| Max output | Not reported | Not reported |
| Input | text | text |
| Output | audio | audio |
When to choose each
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
Sonic 3.6
Best for
- Multilingual speech for voice agents
- Indian-language and Hinglish speech
Watch out for
- Credit-based plans only (1 credit per character); the recorded rate is the Pro plan overage of $65 per 1M characters, falling to $38 on the Scale plan
- Cartesia does not state latency figures for this version
Eleven v4 vs Sonic 3.6 FAQ
Which is cheaper, Eleven v4 or Sonic 3.6?
Sonic 3.6 is 19% cheaper per character than Eleven v4, $0.000065 vs $0.00008.
Which scores higher on independent benchmarks, Eleven v4 or Sonic 3.6?
Eleven v4 leads on Artificial Analysis Speech Arena, 1326 Elo to 1286 Elo. Scores are independent evaluations, not either vendor's launch figures.
Related comparisons
Production evaluation
Choosing between Eleven v4 and Sonic 3.6?
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