ComfyUI Node
Qwen TTS Voice Design Node
A ComfyUI node in Vantage/Audio/Qwen3 TTS with 14 inputs and 1 output.
Qwen TTS Voice Design Node
- model
- tokenizer
- AUDIO
◄text—►
◄instruct—►
◄languageAuto►
◄deviceinherit►
◄attentioninherit►
◄seed0►
◄max_new_tokens2048►
◄top_p0.80►
◄top_k20►
◄temperature1.00►
◄repetition_penalty1.05►
◄unload_model_after_generatefalse►
CategoryVantage/Audio/Qwen3 TTS
Inputs (14)
| Name | Type | Default | Description |
|---|---|---|---|
| model | QWEN_TTS_MODEL | — | |
| tokenizer | QWEN_TTS_TOKENIZER | — | |
| text | STRING | — | |
| instruct | STRING | — | |
| language | COMBO | Auto | 11 options: Auto, Chinese, English, Japanese, Korean, French, +5 |
| device | COMBO | inherit | 5 options: inherit, auto, cuda, mps, cpu |
| attention | COMBO | inherit | 6 options: inherit, auto, sage_attn, flash_attn, sdpa, eager |
| seedopt | INT | 0 | — |
| max_new_tokensopt | INT | 2048 | — |
| top_popt | FLOAT | 0.80 | — |
| top_kopt | INT | 20 | — |
| temperatureopt | FLOAT | 1.00 | — |
| repetition_penaltyopt | FLOAT | 1.05 | — |
| unload_model_after_generateopt | BOOLEAN | false | Unload model from memory after generation |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| AUDIO | AUDIO | — |