Data read October 27, 2026 · Coval, Speko, Artificial Analysis, Cekura
Cartesia alternatives in 2026: 21 multilingual TTS APIs compared
Cartesia's best-ranked model is Cartesia Sonic 3.6, rank 14 with a P95 of 374.4 ms. 13 other ranked models have a lower P95; the lowest is Alibaba Qwen3 TTS Fast.
This is a data comparison, not a recommendation: speed is only one of the measures, so quality, accuracy and price are in the columns beside it.
How the rows are ordered. Cartesia's own models first, then the ranked alternatives. Models are ordered by speed tier, then by P95 (the slow turn). Two P95 values within 10 ms share a rank, and tied models are listed by median latency, a rule applied to every tie.
Cartesia models in the database
Ranked models keep their rank; the others are not tagged multilingual by Coval or not measured by it.
| Rank | Model | Provider | Speed tier From the Coval P95 | P95 (ms) 1 turn in 20 is slower | Listener Elo Artificial Analysis, with 95% interval | Price per 1M characters List price; a range when sources differ | Best for Computed: where the model leads among ranked models |
|---|---|---|---|---|---|---|---|
| 14 |  Cartesia Sonic 3.6 | Cartesia | Visible delay | 374.4 | 1,292 ±14 | $39.20 to $50 | Rated for knowledge sharing and assistants |
| 15 |  Cartesia Sonic 3.5 | Cartesia | Visible delay | 390.5 | 1,190 ±11 | $39.20 to $50 | Most languages tested |
| not ranked |  Cartesia Sonic 3 | Cartesia | n/a | n/a | 1,054 ±11 | $49 to $50 |
Ranked alternatives (21 models)
| Rank | Model | Provider | Speed tier From the Coval P95 | P95 (ms) 1 turn in 20 is slower | Listener Elo Artificial Analysis, with 95% interval | Price per 1M characters List price; a range when sources differ | Best for Computed: where the model leads among ranked models |
|---|---|---|---|---|---|---|---|
| 1 = |  Gradium TTS | Gradium | Headroom | 104.3 | 1,174 ±13 | $36 to $57.80 | Lowest median latency |
| 1 = |  Inworld Realtime TTS-2 Flash | Inworld | Headroom | 100.9 | 1,218 ±13 | $10.40 to $15 | Rated for knowledge sharing and assistants |
| 1 = |  Alibaba Qwen3 TTS Fast | Alibaba (Qwen) | Headroom | 98.8 | n/a | $10 | Lowest P95 latency |
| 4 |  Alibaba Qwen3 TTS 1.7b | Alibaba (Qwen) | Fast enough | 136.6 | n/a | n/a | Rated for voice agents |
| 5 |  SpeechifyAI Simba 3.0 | Speechify | Fast enough | 149.4 | 1,116 ±11 | $6.60 to $10 | Lowest list price |
| 6 |  Inworld Realtime TTS-2 | Inworld | Fast enough | 194.7 | 1,267 ±14 | $15 to $25 | Best reading accuracy (Cekura) |
| 7 |  Palabra TTS v1 | Palabra | Fast enough | 214.3 | n/a | $12 | Rated for voice agents |
| 8 |  ElevenLabs Eleven v4 Turbo | ElevenLabs | Fast enough | 238.6 | 1,333 ±16 | $40 | Highest listener Elo; Highest naturalness (Speko) |
| 9 |  ElevenLabs Flash v2.5 | ElevenLabs | Fast enough | 255.1 | 1,076 ±11 | $40 to $50 | Rated for knowledge sharing and assistants |
| 10 = |  Soniox TTS RT v1 | Soniox | Visible delay | 319.0 | n/a | $13 to $14.10 | Visible delay tier, no measure where it leads |
| 10 = |  Soniox TTS Real-Time v2 | Soniox | Visible delay | 324.7 | 1,180 ±11 | $13 to $14.81 | Rated for knowledge sharing and assistants |
| 12 = |  Murf AI Falcon 2 | Murf AI | Visible delay | 363.5 | 1,154 ±12 | $10 | Rated for knowledge sharing and assistants |
| 12 = |  Rime Mist v3 | Rime | Visible delay | 359.9 | n/a | $30 | Visible delay tier, no measure where it leads |
| 16 = |  Deepdub PhantomZ 4.0 | Deepdub | Visible delay | 430.0 | n/a | n/a | Visible delay tier, no measure where it leads |
| 16 = |  Rime Coda | Rime | Visible delay | 428.1 | 1,054 ±12 | $50 | Rated for knowledge sharing and assistants |
| 18 |  SpaceXAI Grok TTS | xAI | Visible delay | 473.9 | 1,155 ±13 | $15 | Rated for knowledge sharing and assistants |
| 19 |  Fish Audio S2.1 Pro | Fish Audio | Visible delay | 533.0 | 1,184 ±13 | $15 | Rated for knowledge sharing and assistants |
| 20 |  Smallest.ai Lightning V3.1 Pro (Jul 2026) | Smallest.ai | Visible delay | 703.3 | 1,169 ±12 | $19.50 | Rated for knowledge sharing and assistants |
| 21 |  Alibaba Qwen3 TTS Flash Realtime | Alibaba (Qwen) | Visible delay | 785.8 | n/a | $13 | Visible delay tier, no measure where it leads |
| 22 |  Google Chirp 3: HD | Visible delay | 917.2 | 1,061 ±12 | $30 | Rated for knowledge sharing and assistants | |
| 23 |  Fish Audio S2.1 Pro Free | Fish Audio | Visible delay | 1,137.9 | n/a | n/a | Visible delay tier, no measure where it leads |
Frequently asked questions
- What are the best alternatives to Cartesia for multilingual real-time TTS?
- In ranking order, the first ranked alternatives are Gradium TTS, Inworld Realtime TTS-2 Flash, Alibaba Qwen3 TTS Fast, Alibaba Qwen3 TTS 1.7b, and SpeechifyAI Simba 3.0. Speed order only: compare listener Elo and price before choosing.
- Is any TTS API faster than Cartesia?
- 13 ranked models have a lower P95 than Cartesia Sonic 3.6. The fastest is Alibaba Qwen3 TTS Fast at 98.8 ms.
- Are the alternatives cheaper than Cartesia?
- Prices are in the column when a benchmark printed one, and 3 ranked models have none. See the pricing page for the cheapest first.
- How were these alternatives chosen?
- Every ranked model that is not Cartesia's is listed, in the ranking order. Nothing is filtered by price, quality or vendor.