whatmodel.sh / tts / MiniMax Speech-02-HD

MiniMax Speech-02-HD

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

MiniMax Speech-02-HD is a closed-weights text-to-speech model from MiniMax, released March 2025. Licence: proprietary.

62.6±2.7#28 of 118 text-to-speech modelsfrom 3 benchmark results across 2 sources

Score within the Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot comparison set only.

$ how it compares

The text-to-speech models ranked either side of MiniMax Speech-02-HD, out of 148 with a score, then more from MiniMax.

#modelscorewhereRAMreleased
34GLM-TTS Z.ai63±2local~8.9GB2025-12
35Sonic 3.5 Cartesia63±2comparison only2026-05
36MiniMax Speech-02-HD MiniMax63±3api $100.00/M chars2025-03
37Fish Audio S2 Pro Fish Audio63±1api $15.00/M chars +local~11GB2026-03
38Eleven Flash v2.5 ElevenLabs63±2api $50.00/M chars2024-12
more from MiniMax
15MiniMax Speech 2.8 Turbo MiniMax66±3api $60.00/M chars2026-02
19MiniMax Speech 2.8 HD MiniMax66±2api $100.00/M chars2026-02
51MiniMax Speech-02-Turbo MiniMax60±3api $60.00/M chars2025-04

see all text-to-speech models · find the ones that fit your machine

$ what it's good at

other tests3 not in a task · 97% of its score
testresultvs the others in its rundifficultycountswhy
Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot1260.7 ✎▁···▁▃··█▄▁▇ beats 14/18avg 1078▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 68%typical
EmergentTTS · 6-category preference36.8% ✎▅▅▃▃█▅▅▃·▃·▅ beats 11/16hard · avg 30%▇▇▇▇▇▇▇▇▇ 24%at its level
EmergentTTS · 6-category WER10.1% ✎▂▄··▂▄▄··█▄▆ beats 16/16avg 16%▇▇ 5.2%blunt

$ where to run it

Through an API

api · $100.00/M chars · platform.minimax.io ↗ · price sourceUSD per 1M input characters, HD model. No tokens-to-characters conversion. Product: Pay-as-you-go text-to-audio. API model: speech-02-hd. Settings: mode=T2A, quality=HD. Price basis: published rate. Currency: USD.

$ how the score was worked out · The Humanness Index by Vapi ·… 68%, EmergentTTS-Eval six-category… 29%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
The Humanness Index by Vapi · September 2026 public standings1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 68%CC BY 4.0
EmergentTTS-Eval six-category leaderboard2▇▇▇▇▇▇▇▇▇ 29%Apache-2.0
prior · pull toward the middle▇ 3%
weighted average= 62.6

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json