Lab · compact cloning LM
OuteTTS-1.0 (0.6B): specs & requirements
OuteTTS-1.0-0.6B is a Qwen3-based speech LM with one-shot voice cloning from ~10 seconds of reference audio, across 12 languages. Official ONNX exports exist, but nobody has demonstrated the full 1.0 pipeline in a browser yet — wiring it here would be a first, which is exactly why it's in the Lab.
Lab tier — heavy/frontier model for enthusiast hardware
OuteTTS-1.0 (0.6B) support is being wired into the tool. Its specs and hardware verdicts are below — meanwhile, the live models are ready to use, and the compatibility check already tells you whether your device could run OuteTTS-1.0 (0.6B).
Specifications
| Developer | OuteAI |
|---|---|
| Parameters | 600 million |
| Download size | 582 MB (q4f16) |
| Output | 24.0 kHz mono |
| Voices | 1 |
| Languages | English, Arabic, Chinese, Dutch, French, German, Italian, Japanese, Korean, Lithuanian, Russian, Spanish |
| Voice cloning | Yes |
| Streaming | No |
| License | Apache-2.0 — The 1.0-0.6B release (Qwen3 base) is fully Apache-2.0 — unlike the older 0.2/0.3 OuteTTS models, which are non-commercial. |
| Speed | Usable on WebGPU; autoregressive generation makes CPU runs impractical. |
| Runtime | WebGPU required |
Source: OuteAI/OuteTTS-1.0-0.6B-ONNX
Strengths
- +One-shot cloning from a short reference clip
- +Fully permissive Apache-2.0 (rare for cloning-capable LMs)
- +Compact 582 MB q4f16 download for a 0.6B model
Limitations
- −Browser pipeline unproven — official exports, no public demo yet
- −Strict sampling requirements; quality degrades if parameters drift
OuteTTS-1.0 (0.6B) — questions
Is OuteTTS-1.0 (0.6B) free to use?
Yes. The weights are Apache-2.0 and they run on your device. The 1.0-0.6B release (Qwen3 base) is fully Apache-2.0 — unlike the older 0.2/0.3 OuteTTS models, which are non-commercial.
What hardware does OuteTTS-1.0 (0.6B) need?
A WebGPU-capable browser and GPU (Chrome/Edge, Safari 26+, Firefox 141+) plus about 8 GB of RAM. It will not run on CPU at usable speed.
Is my text private when using OuteTTS-1.0 (0.6B) here?
Yes. The model executes inside your browser tab — the text you synthesize is never transmitted to a server.
Can I use OuteTTS-1.0 (0.6B) output commercially?
Review the current model, voice, and source-dataset terms before commercial use. The catalog lists Apache-2.0 for this model and notes: The 1.0-0.6B release (Qwen3 base) is fully Apache-2.0 — unlike the older 0.2/0.3 OuteTTS models, which are non-commercial.