Data read October 27, 2026
TTS APIs rated for voice agents in 2026: 13 models
13 models have models rated for voice agents. 9 of the 13 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 4 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 |
| 1 = | | Alibaba (Qwen) | Headroom | 98.8 | n/a | $10 |
| 4 | | Alibaba (Qwen) | Fast enough | 136.6 | n/a | n/a |
| 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 |
| 7 | | Palabra | Fast enough | 214.3 | n/a | $12 |
| 8 | | ElevenLabs | Fast enough | 238.6 | 1,333 ±16 | $40 |
| 9 | | ElevenLabs | Fast enough | 255.1 | 1,076 ±11 | $40 to $50 |
| not ranked | | Deepgram | Fast enough | 267.0 | 1,052 ±13 | $45 |
| not ranked | | Fish Audio | Fast enough | 209.3 | 1,245 ±14 | $15 |
| not ranked | | Fluxions | Headroom | 75.0 | n/a | $9 |
| not ranked | | Speechify | Fast enough | 161.1 | 1,249 ±12 | $6.60 to $10 |
Frequently asked questions
- Which TTS APIs have models rated for voice agents?
- Starting with the best-ranked, Gradium TTS, Inworld Realtime TTS-2 Flash, Alibaba Qwen3 TTS Fast, and Alibaba Qwen3 TTS 1.7b; 13 in total.
- How many are ranked?
- 9 of the 13 are in the multilingual speed ranking; the others are not tagged multilingual by Coval or are not measured by it.