whatmodel.sh / tts / MiniMax Speech 2.8 HD
MiniMax Speech 2.8 HD
↗ share
+ contribute
MiniMax Speech 2.8 HD is a closed-weights text-to-speech model from MiniMax, released February 2026. Licence: proprietary.
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 2.8 HD, out of 148 with a score, then more from MiniMax.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 17 |  LongCat-AudioDiT-1B | 66±4 | local | ||
| 18 |  OmniVoice | 66±2 | local | ||
| 19 | MiniMax Speech 2.8 HD | 66±2 | api $100.00/M chars | ||
| 20 |  dots.tts SOAR 2B | 65±2 | local | ||
| 21 |  Eleven v3 | 65±1 | api $100.00/M chars | ||
| more from MiniMax | |||||
| 15 | MiniMax Speech 2.8 Turbo | 66±3 | api $60.00/M chars | ||
| 36 | MiniMax Speech-02-HD | 63±3 | api $100.00/M chars | ||
| 51 | MiniMax Speech-02-Turbo | 60±3 | api $60.00/M chars | ||
see all text-to-speech models · find the ones that fit your machine
$ what it's good at
other tests8 not in a task · 99% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CN-NewsTTS v0.1 · strict pronunciation accuracy | 54.8% ✎ | █··▄·█··▄··▄ beats 3/6 | 21% |
| CV3-Eval hard-en · WER | 7.4% ✎ | █····█·█████ beats 3/6 | 18% |
| CV3-Eval hard-en · DNSMOS (predicted quality) | 3.80 / 5 ✎ | █·▄·▄▄·▄···▄ beats 2/6 | 14% |
| CV3-Eval hard-zh · CER | 7.4% ✎ | ▄▄▄···▄··▄·█ beats 6/6 | 13% |
| CV3-Eval multilingual voice clone · 9-language mean error | 4.9% ✎ | ▄···▄··▄··██ beats 5/6 | 12% |
| CV3-Eval protocol, 7 added languages · mean WER (Qwen-Audio 3.0 extension) | 2.8% ✎ | █··█······██ beats 3/3 | 8.7% |
| CV3-Eval hard-en · speaker similarity (ERes2Net SV) | 72.1% ✎ | ▄·····▄▄·▄▄█ beats 3/6 | 6.6% |
| Voice-First Nation · 10 Indian languages overall pairwise preference (Bradley–Terry) | 993.9 ✎ | ▄····▄·····█ beats 1/3 | 5.1% |
$ where to run it
Through an API
api · $100.00/M chars · platform.minimax.io ↗ · price sourceUSD per 1M input characters (MiniMax lists $100 / M characters; sync T2A and async T2A same rate). Product: MiniMax pay-as-you-go text-to-audio (T2A). API model: speech-2.8-hd. Settings: pay-as-you-go. Price basis: published list rate. Exclusions: voice design/cloning fees. Currency: USD. MiniMax official pay-as-you-go pricing page, observed 2026-09-25.
$ how the score was worked out · Qwen-Audio-3.0-TTS technical… 72%, CN-NewsTTS Bench v0.1 21%, Preferences of a Voice-First… 5%
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
| source | results | share of the score |
|---|---|---|
 Qwen-Audio-3.0-TTS technical report | 6 | 72% |
 CN-NewsTTS Bench v0.1 | 1 | 21% |
| Preferences of a Voice-First Nation: pairwise human evaluation of TTS in 10 Indian languages (arXiv:2604.21481v2) | 1 | 5% |
| prior · pull toward the middle | 1% | |
| weighted average | = 65.6 |
$ improve this page
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