Data read October 27, 2026
TTS APIs rated for assistants in 2026: 69 models
69 models have models rated for assistants. 15 of the 69 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.
Artificial Analysis assigns these use-case categories; the site does not decide them.
How the rows are ordered. Ranked models first, in ranking order (speed tier, then P95, ties by median); the 54 models without a rank follow alphabetically.
| 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 |
|---|---|---|---|---|---|---|
| 1 = | | Gradium | Headroom | 104.3 | 1,174 ±13 | $36 to $57.80 |
| 1 = | | Inworld | Headroom | 100.9 | 1,218 ±13 | $10.40 to $15 |
| 5 | | Speechify | Fast enough | 149.4 | 1,116 ±11 | $6.60 to $10 |
| 6 | | Inworld | Fast enough | 194.7 | 1,267 ±14 | $15 to $25 |
| 8 | | ElevenLabs | Fast enough | 238.6 | 1,333 ±16 | $40 |
| 9 | | ElevenLabs | Fast enough | 255.1 | 1,076 ±11 | $40 to $50 |
| 10 = | | Soniox | Visible delay | 324.7 | 1,180 ±11 | $13 to $14.81 |
| 12 = | | Murf AI | Visible delay | 363.5 | 1,154 ±12 | $10 |
| 14 | | Cartesia | Visible delay | 374.4 | 1,292 ±14 | $39.20 to $50 |
| 15 | | Cartesia | Visible delay | 390.5 | 1,190 ±11 | $39.20 to $50 |
| 16 = | | Rime | Visible delay | 428.1 | 1,054 ±12 | $50 |
| 18 | | xAI | Visible delay | 473.9 | 1,155 ±13 | $15 |
| 19 | | Fish Audio | Visible delay | 533.0 | 1,184 ±13 | $15 |
| 20 | | Smallest.ai | Visible delay | 703.3 | 1,169 ±12 | $19.50 |
| 22 | | Visible delay | 917.2 | 1,061 ±12 | $30 | |
| not ranked | | Alibaba (Qwen) | n/a | n/a | 1,269 ±13 | $19.30 |
| not ranked | | Alibaba (Qwen) | n/a | n/a | 1,306 ±15 | $19.30 |
| not ranked | | Alibaba (Qwen) | n/a | n/a | 929 ±13 | $13.80 |
| not ranked | | Alibaba (Qwen) | n/a | n/a | 939 ±14 | $10 to $11 |
| not ranked | | Amazon | n/a | n/a | 1,063 ±11 | $30 |
| not ranked | | Amazon | n/a | n/a | 1,029 ±13 | $100 |
| not ranked | | Amazon | n/a | n/a | 894 ±14 | $16 |
| not ranked | | Amazon | n/a | n/a | 829 ±16 | $4 |
| not ranked | | Async | n/a | n/a | 1,049 ±11 | $10.10 |
| not ranked | | Async | n/a | n/a | 1,137 ±12 | $10.10 |
| not ranked | | Async | n/a | n/a | 1,127 ±12 | $20.20 |
| not ranked | | Bland AI | n/a | n/a | 1,033 ±12 | $40 |
| not ranked | | Boson AI | n/a | n/a | 1,045 ±13 | $15 |
| not ranked | | BreezeBlue | n/a | n/a | 1,227 ±13 | $34 |
| not ranked | | Cartesia | n/a | n/a | 1,054 ±11 | $49 to $50 |
| not ranked | | Deepgram | Fast enough | 267.0 | 1,052 ±13 | $45 |
| not ranked | | ElevenLabs | n/a | n/a | 1,184 ±10 | $100 |
| not ranked | | ElevenLabs | n/a | n/a | 1,213 ±12 | $50 |
| not ranked | | ElevenLabs | n/a | n/a | 1,333 ±16 | $80 |
| not ranked | | ElevenLabs | n/a | n/a | 1,093 ±10 | $100 |
| not ranked | | ElevenLabs | n/a | n/a | 1,093 ±10 | $50 |
| not ranked | | Fish Audio | Fast enough | 209.3 | 1,245 ±14 | $15 |
| not ranked | | Fish Audio | n/a | n/a | 1,084 ±11 | $15 |
| not ranked | | Fish Audio | n/a | n/a | 1,108 ±12 | $15 |
| not ranked | | n/a | n/a | 1,087 ±11 | $9.20 | |
| not ranked | | n/a | n/a | 1,067 ±12 | $9.20 | |
| not ranked | | n/a | n/a | 1,040 ±12 | $18.30 | |
| not ranked | | n/a | n/a | 1,212 ±11 | $18.30 to $33.30 | |
| not ranked | | n/a | n/a | 1,280 ±14 | $15 to $16.50 | |
| not ranked | | n/a | n/a | 1,256 ±13 | $10 to $11 | |
| not ranked | | Gradium | n/a | n/a | 1,151 ±12 | $47.20 |
| not ranked | | Hume AI | n/a | n/a | 1,056 ±12 | $87.50 to $100 |
| not ranked | | Hume AI | n/a | n/a | 1,036 ±11 | $87.50 to $100 |
| not ranked | | KugelAudio | n/a | n/a | 924 ±13 | n/a |
| not ranked | | Maya Research | n/a | n/a | 1,065 ±13 | n/a |
| not ranked | | Maya Research | n/a | n/a | 1,047 ±13 | n/a |
| not ranked | | Maya Research | n/a | n/a | 1,048 ±12 | n/a |
| not ranked | | Microsoft | n/a | n/a | 1,142 ±11 | $22 |
| not ranked | | MiniMax | n/a | n/a | 1,126 ±11 | $100 |
| not ranked | | MiniMax | n/a | n/a | 1,120 ±11 | $60 |
| not ranked | | MiniMax | n/a | n/a | 1,174 ±10 | $100 |
| not ranked | | MiniMax | n/a | n/a | 1,153 ±10 | $60 |
| not ranked | | MiniMax | n/a | n/a | 1,108 ±11 | $100 |
| not ranked | | MiniMax | n/a | n/a | 1,089 ±11 | $60 |
| not ranked | | Mistral | n/a | n/a | 1,221 ±14 | $13 |
| not ranked | | Mistral | n/a | n/a | 1,088 ±12 | $16 |
| not ranked | | Neuphonic | n/a | n/a | 940 ±14 | $20.80 |
| not ranked | | OpenAI | n/a | n/a | 1,073 ±12 | $191.60 |
| not ranked | | OpenAI | n/a | n/a | 1,088 ±10 | $15 |
| not ranked | | OpenAI | n/a | n/a | 1,106 ±11 | $30 |
| not ranked | | Smallest.ai | n/a | n/a | 1,020 ±12 | $17.50 to $25 |
| not ranked | | Speechify | Fast enough | 161.1 | 1,249 ±12 | $6.60 to $10 |
| not ranked | | StepFun | n/a | n/a | 1,135 ±11 | $40 |
| not ranked | | StepFun | n/a | n/a | 1,212 ±13 | $85 |
Frequently asked questions
- Which TTS APIs have models rated for assistants?
- Starting with the best-ranked, Gradium TTS, Inworld Realtime TTS-2 Flash, SpeechifyAI Simba 3.0, and Inworld Realtime TTS-2; 69 in total.
- How many are ranked?
- 15 of the 69 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.