Vidu Q3 Text To Video (Official Edition)
Vidu Q3 Turbo text-to-video, Official edition — the full-speed text route with off-peak savings
- api_config
- video
- request_id
- urls
Turbo text-to-video with the Official knob set is the sweet spot for anyone who generates a lot of text-driven clips: it's the fastest Vidu tier, it has seed control for reproducibility, and it keeps the off_peak discount mode for batch work. Pro gets you the polish; Turbo gets you the volume.
What this node does
BizyTRD_Vidu_Q3_Turbo_Official_T2V_API turns a prompt into a clip on the Turbo tier with the full Official edition - seed, audio, and off-peak included. Like every BizyTRD node, it's cloud execution: flip BizyAirPlus ON, Turbo runs on BizyAir's servers, and ComfyUI streams back the result. Your GPU never touches the model.
The inputs that matter
prompt- up to 5000 characters; everything lives here.resolution- 720P default.duration- 1–16 seconds.aspect_ratio- 16:9 default.seed- 0 default, -1 = random. Lock it to reproduce a take.audio- on by default: Q3 generates dialogue and SFX.off_peak- the money-saver: lower point cost, completion within 48 hours, auto-cancel-and-refund on timeout. Perfect for overnight batches of clips; wrong choice when you're waiting on a client.
Shared BizyAir options: api_config (endpoint/key override) and skip_error (placeholder instead of abort).
Outputs
video- the finished clip, wired into preview or save.request_id- cloud task ID.urls- hosted result URLs.
How to install
cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt
Restart ComfyUI (or install "BizyAirPlus" via ComfyUI Manager). Deps: bizyair-cloudberry and bizytrd; no model files. First run: account at bizyair.ai → API key → enter when the ON button prompts (or Settings > BizyAirPlus > API Key; key at ~/.BizyAirPlus/apikey.ini).
Where people get burned
The off_peak refresh trap is the classic: enable it, then sit refreshing ComfyUI like it's a normal queue. It's not - 48 hours is the deal, and that's the point (cheaper). Also remember T2V is the weakest mode for subject consistency; Turbo T2V especially. If faces drift, anchor a first frame with the I2V or FLF nodes rather than fighting the prompt.
Pack note: BizyAir's malware reputation is leftover from a 2025 ComfyUI Manager auto-install bug - legitimate SiliconFlow pack, removed properly with pip uninstall bizyair-cloudberry bizytrd.
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | Vincenzo video prompt word, Max 5000 characters | |
| resolution | COMBO | 720P | Output resolution, default 720P |
| duration | INT | 51–16 | Video length (seconds) , range 1-16, default 5 |
| aspect_ratioopt | COMBO | 16:9 | Output video aspect ratio, 16:9 by default |
| seedopt | INT | 0-1–2147483647 | Random Seed, -1 indicates automatic random generation |
| audioopt | BOOLEAN | true | Enables the generation of audio with video (including dialogue and sound effects) , only Q3 model support |
| off_peakopt | BOOLEAN | false | Enable trough mode to consume lower points, complete within 48 hours, timeout automatically cancels and refunds |
| api_configopt | BIZYAIR_OPENAPI_CONFIG | 单次请求覆盖base_url和api_key | |
| skip_erroropt | BOOLEAN | false | 开启后遇到错误不中断工作流,输出对应类型的错误占位符 |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |
| request_id | STRING | — |
| urls | STRING | — |