CartesiaMiniMaxAudio matchupSpeech output
Sonic 3.6vs.MiniMax Speech 2.8
Which model is better?
Sonic 3.6 does better in independent tests.
Pick Sonic 3.6 for
Pick MiniMax Speech 2.8 for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
VoiceArena
Sonic 3.6 aheadSonic 3.6 is clearly ahead in 3 of 3 subtests, MiniMax Speech 2.8 in none.
VoiceArena TTS Elo: Mexican Spanish
95 % ranges 1,065 to 1,097 and 769 to 817, as of October 28, 2026
VoiceArena TTS Elo: Brazilian Portuguese
95 % ranges 1,062 to 1,094 and 841 to 861, as of October 28, 2026
VoiceArena TTS Elo: Hindi
95 % ranges 1,081 to 1,107 and 931 to 951, as of October 28, 2026
Source · latest October 28, 2026
Artificial Analysis
Sonic 3.6 aheadSonic 3.6 is clearly ahead in 2 of 2 subtests, MiniMax Speech 2.8 in none.
Artificial Analysis Controlled Voice Arena Elo
95 % ranges 1,135 to 1,159 and 1,012 to 1,036, as of November 3, 2026
Artificial Analysis Provider Voice Arena preference
95 % ranges 1,280 to 1,306 and 1,163 to 1,183, as of November 3, 2026
Source · latest November 3, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| Sonic 3.6 | MiniMax Speech 2.8 | |
|---|---|---|
| Task | Speech output | Speech output |
| Developer | Cartesia | MiniMax |
| Status | Active | Active |
| Released | n/a | January 23, 2026 |
| Weights | proprietary | proprietary |
| Runs locally | no | no |
| Price | n/a | speech-2.8-hd costs $100 per 1M characters, synchronous and asynchronous |
| Popularity (out of 100) | 39 | 0 |