All tasks / Models
Speech 2.8 HD
MiniMax's high-definition text-to-speech model line with voice cloning, available via API and MiniMax Audio.
- MiniMax
- Closed
- Released 14 Feb 2026
Where it ranks
- BText to speechModels · Cloud · 1,174 on Artificial Analysis Speech Arena: Text to Speech (provider voices)
Facts
- Vendor
- MiniMax
- Family
- MiniMax Speech
- Released
- 14 Feb 2026
- Weights
- Closed: API and apps only
- License
- proprietary
- API price
- $100 per 1M characters (Artificial Analysis normalised price for Speech 2.8 HD (USD per 1M characters).)
- Last verified
- 18 Oct 2026
Price from artificialanalysis.ai, 18 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Speech Arena: Text to Speech (provider voices) | 1,174 | Speech 2.8 HDProvider Voice Arena, all categories and accents; rank 22 of 96, 5043 appearances, 95% CI ±11; raw value from the page's embedded data (shown rounded); no last-updated date shown, so as_of = retrieval date. | 18 Oct 2026 | artificialanalysis.ai |
| TTS Arena V2 (Hugging Face Space) | 1,535 | MiniMax Speech 2.8 HDBradley-Terry rating ±30, 503 votes, board rank 7 (the board sorts by the lower bound of the interval); read from the arena's public JSON endpoint, which feeds https://huggingface.co/spaces/TTS-AGI/TTS-Arena-V2. | 18 Oct 2026 | tts-agi-tts-arena-v2.hf.space |
Where you can use it
No tracked tool lists this model yet.
Earlier versions
- Speech 2.6 HD · 30 Oct 2025
- Speech-02-HD · 31 Mar 2025