ElevenLabs vs Cartesia
| Measure | ElevenLabsEleven v4 | CartesiaSonic 3.6 |
|---|---|---|
| Score (of 100) | 86 (better) | 83 |
| Rank overall | #2 (better) | #3 |
| Boards listed on | 2 of 3 | 2 of 3 |
| Price per 1M characters | $80.00 | $49.00 (better) |
| Coval time to first audio (p50) | — | 288 ms |
| Speko time to first audio (p50) | 844 ms | 120 ms (better) |
| Humanness Index time to first audio (median) | 671 ms | 101 ms (better) |
| Open weights | No | No |
"—" means the source has not published a figure for that model. Latency sources are never averaged together.
Short answer
Eleven v4 scores higher (86 to 83), but Sonic 3.6 costs 39% less. The latency sources split on who is faster.
Quality
Eleven v4 (ElevenLabs) scores 86, 3 points above Sonic 3.6 (Cartesia) at 83. They rank #2 and #3 of 121 models.
Both models are listed on 2 of 3 boards, so neither score has the full set behind it.
Price
Sonic 3.6 lists at $49.00 per 1M characters. Eleven v4 lists at $80.00, 63% more.
ElevenLabs also has Eleven v4 Turbo at $40.00, which scores 70.
Speed
Coval has no figure for Eleven v4, so ElevenLabs is represented by Eleven v4 Turbo. Coval: Eleven v4 Turbo at 191 ms, Sonic 3.6 at 288 ms. ElevenLabs is 1.5× faster.
Speko: Sonic 3.6 at 120 ms, Eleven v4 at 844 ms. Cartesia is 7× faster.
Humanness Index: Sonic 3.6 at 101 ms, Eleven v4 at 671 ms. Cartesia is 6.6× faster.
The sources do not agree on who is faster. They use different harnesses, regions and sample sizes, so treat speed as unsettled.
Model lineups
| Model | Rank | Score | Price$ / 1M chars | Covalp50 ms | Spekop50 ms | Humanness Indexmedian ms |
|---|---|---|---|---|---|---|
| Eleven v4ElevenLabs | #2 | 86 | $80.00 | — | 844 ms | 671 ms |
| Eleven v3ElevenLabs | #8 | 74 | $100.00 | — | — | 758 ms |
| v3 ConversationalElevenLabs | #11 | 71 | $50.00 | — | 266 ms | — |
| Eleven v4 TurboElevenLabs | #12 | 70 | $40.00 | 191 ms | 205 ms | 158 ms |
| Turbo v2.5ElevenLabs | #52 | 37 | $50.00 | — | — | — |
| Sonic 3.6Cartesia | #3 | 83 | $49.00 | 288 ms | 120 ms | 101 ms |
| Sonic 3.5Cartesia | #10 | 73 | $49.00 | — | 121 ms | 128 ms |
| Sonic 3Cartesia | #60 | 31 | $49.00 | — | — | 166 ms |
| Sonic 2Cartesia | #110 | 3 | — | — | — | 159 ms |
| SonicCartesia | #120 | 0 | — | — | — | 116 ms |
ElevenLabs has 3 more models on the provider page.
ElevenLabs has 8 models on the board, ranging from 86 down to 7; 4 sit in the top 20. Cartesia has 5 models, from 83 to 0; 2 in the top 20.
Neither provider lists an open-weights model.
Questions people ask
- Is Cartesia better than ElevenLabs?
- On our score, no. Eleven v4 (ElevenLabs) scores 86, 3 points above Sonic 3.6 (Cartesia) at 83. They rank #2 and #3 of 121 models. Sonic 3.6 lists at $49.00 per 1M characters. Eleven v4 lists at $80.00, 63% more.
- Which is cheaper, ElevenLabs or Cartesia?
- Sonic 3.6 lists at $49.00 per 1M characters. Eleven v4 lists at $80.00, 63% more. ElevenLabs also has Eleven v4 Turbo at $40.00, which scores 70.
- Which is faster, ElevenLabs or Cartesia?
- Coval has no figure for Eleven v4, so ElevenLabs is represented by Eleven v4 Turbo. Coval: Eleven v4 Turbo at 191 ms, Sonic 3.6 at 288 ms. ElevenLabs is 1.5× faster. Speko: Sonic 3.6 at 120 ms, Eleven v4 at 844 ms. Cartesia is 7× faster. Humanness Index: Sonic 3.6 at 101 ms, Eleven v4 at 671 ms. Cartesia is 6.6× faster. The sources do not agree on who is faster. They use different harnesses, regions and sample sizes, so treat speed as unsettled.
Where the numbers come from
Quality is the 0–100 score from our rank aggregation of 3 public boards: artificialanalysis.ai (Speech Arena Elo), voicearena.com (arena Elo, US English) and humannessindex.vapi.ai (humanness score 0–100). Prices are dollars per 1M characters, taken from Artificial Analysis and provider pages; per-minute prices are converted to the same unit. Latency is time to first audio (the wait before sound starts), copied from Coval, Speko and Humanness Index and never averaged across them. We do not measure anything ourselves. How latency is sourced.