whatmodel.sh / tts / MGM-Omni-TTS 2B (0927)

MGM-Omni-TTS 2B (0927)

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

MGM-Omni-TTS 2B (0927) is an open-weights text-to-speech model from MGM-Omni (CUHK), released September 2025. Its published size is 2.15B parameters. Licence: apache-2.0.

Catalogue identity notes

wcy1122/MGM-Omni-TTS-2B-0927 (2,151,311,872 parameters per HF safetensors metadata).

58.7±4.3comparison only: no current run option verifiedfrom 4 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 MGM-Omni-TTS 2B (0927), out of 148 with a score.

#modelscorewhereRAMreleased
52TTS-1 OpenAI59±4api $15.00/M chars2023-11
53MaskGCT Amphion (CUHK-Shenzhen)59±1local~6.6GB2024-10
54MGM-Omni-TTS 2B (0927) MGM-Omni (CUHK)59±4comparison only2025-09
55CosyVoice v3 Plus Alibaba Cloud58±4api $26.00/M chars2025-09
56Octave TTS (Octave 1) Hume AI58±4api $50.00/M chars2025-02

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

$ what it's good at

voice cloning▁·▁▁▁▁▁▁██▃▃ #31 of 43 * · 66% of its score · 2 tests

Beats CosyVoice2-0.5B, MaskGCT here, though they rank higher overall. Trails Higgs Audio V2, F5-TTS Base (v0) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Seed-TTS test-en · Whisper large-v3 WER2.5% ✎▃··▃▅··▅·█▃▅ beats 2/11avg 3.6%▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 35%blunt
Seed-TTS test-zh · Paraformer CER1.3% ✎▂··▄··▂▂█·▄▂ beats 2/11avg 2.1%▇▇▇▇▇▇▇▇▇▇▇▇ 31%blunt
other tests2 not in a task · 27% of its score
testresultvs the others in its rundifficultycountswhy
RVCBench LibriTTS clean voice cloning · ECAPA speaker similarity54.0% ✎▃·····▂▂▂▃█▆ beats 6/15medium · avg 46%▇▇▇▇▇▇▇▇▇▇▇ 26%blunt · at its level
RVCBench LibriTTS clean voice cloning · Whisper-medium WER9.0% ✎▂▂▂··▂·▃▃▃█▂ beats 5/15avg 8.6%▇ 0.3%blunt · noisy

$ where to run it

No verified API or download recorded yet · add one

Weights and documentation: huggingface.co/wcy1122 ↗

$ how the score was worked out · UltraEval-Audio (OpenBMB) · T… 66%, RVCBench: Benchmarking Robust… 27%

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
UltraEval-Audio (OpenBMB) · TTS replication results2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%Apache-2.0
RVCBench: Benchmarking Robustness of Voice Cloning Across Modern Audio Generation Models (arXiv:2602.00443v2)2▇▇▇▇▇▇▇▇ 27%CC BY 4.0
prior · pull toward the middle▇▇ 8%
weighted average= 58.7

$ 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