whatmodel.sh / tts / LongCat-AudioDiT-1B

LongCat-AudioDiT-1B

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

LongCat-AudioDiT-1B is an open-weights text-to-speech model from Meituan, released March 2026. Its published size is 1B parameters. Licence: mit.

Catalogue identity notes

Meituan LongCat team's diffusion TTS in the waveform latent space, exact checkpoint meituan-longcat/LongCat-AudioDiT-1B @ 4f4c19434f2e3b991998d0ead6ab2cd056d947b7 (MIT; created 2026-03-30). The 1B row of the LongCat-AudioDiT Seed-benchmark table (card and arXiv:2603.29339v1 Table 1). Distinct from the LongCat LLMs and LongCat-Audio-Codec.

Model information sources

Release date ↗ · Weights licence ↗ · Parameter count ↗

Maker's size label 'LongCat-AudioDiT-1B' (diffusion backbone). HF safetensors header total 1,419,923,649 F32 parameters for the single model.safetensors, which also holds the Wav-VAE and UMT5-base text encoder.

65.8±4.1#11 of 118 text-to-speech modelsfrom 4 benchmark results across 1 source

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 LongCat-AudioDiT-1B, out of 148 with a score.

#modelscorewhereRAMreleased
15MiniMax Speech 2.8 Turbo MiniMax66±3api $60.00/M chars2026-02
16LongCat-AudioDiT-3.5B Meituan66±2local~15GB2026-03
17LongCat-AudioDiT-1B Meituan66±4local~5.7GB2026-03
18OmniVoice k2-fsa (Xiaomi)66±2local~3.3GB2026-03
19MiniMax Speech 2.8 HD MiniMax66±2api $100.00/M chars2026-02

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

$ what it's good at

voice cloning▁·▁▁▁▁▁▁▇█▃▃ #8 of 37 · 93% of its score · 4 tests

Beats Gemini 3.1 Flash TTS Preview here, though they rank higher overall. Trails VoxCPM2, dots.tts SOAR 2B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Seed-TTS test-en · speaker similarity (WavLM-large SV) lab76.2% ✎▁·▂▂▂█▆▃▅··▂ beats 22/30medium · avg 53%▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 36%at its level · small run
Seed-TTS test-zh · speaker similarity (WavLM-large SV) lab81.2% ✎▁·▂··▂·█▄·▃▄ beats 24/25medium · avg 62%▇▇▇▇▇▇▇▇▇▇▇▇▇ 32%small run
Seed-TTS test-en · Whisper large-v3 WER lab1.8% ✎▁···▂▂▂▂█▆▄▂ beats 19/35avg 3.6%▇▇▇▇▇▇ 14%blunt · small run
Seed-TTS test-zh · Paraformer CER lab1.2% ✎▁·▂▁▁▁▂▃▅█·▂ beats 14/30avg 2.1%▇▇▇▇ 11%blunt · small run

$ where to run it

Other ways to run it

  1. Run it as published · needs ~5.68 GB · from huggingface.co/meituan-longcat
    git clone https://github.com/meituan-longcat/LongCat-AudioDiT.git && cd LongCat-AudioDiT && pip install -r requirements.txt && python inference.py --text "Today is a sunny day, let's go for a walk in the park." --output_audio output.wav --model_dir meituan-longcat/LongCat-AudioDiT-1B

Weights and documentation: huggingface.co/meituan-longcat ↗

$ how the score was worked out · meituan-longcat/LongCat-Audio… 93%

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
meituan-longcat/LongCat-AudioDiT-3.5B model card (Seed benchmark table) · lab-reported4▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 93%MIT
prior · pull toward the middle▇▇ 7%
weighted average= 65.8

$ 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