MiniMax Hailuo 2.3 Fast (Channel Edition)
Hailuo 2.3 Fast text-to-video, prompt-only, hosted
- api_config
- video
- request_id
- urls
MiniMax Hailuo 2.3 Fast, text-to-video, channel edition. A prompt in, a video out - no first frame, no image input at all. Hailuo's T2V is well-liked for punchy, physically sensible motion, and the Fast variant exists for when you'd rather have a clip in a couple of minutes than the slow path's marginal quality bump.
This is the node you reach for when you have a video idea and nothing else - no keyframe, no reference footage. You type the shot and it makes it. For a beginner that's genuinely the least intimidating entry into video generation in this pack: there's nothing to prepare, and the model handles all the hard parts.
How it works
Standard BizyTRD cloud wrapper. Prompt goes to the BizyAir channel, Hailuo generates, VIDEO comes back. Two controls matter, and both have the same quirk as the I2V version: resolution supports 768P and 1080p but only 768P is available at 10 seconds - 1080p clips are 6 seconds max. duration defaults to 6 seconds, which is honestly the right starting point.
Prompts accept up to 2000 characters and, per the tooltip, support "conch mirror instruction writing" - the descriptive, camera-move-heavy prompt style from MiniMax's own consumer app. If you've seen Hailuo clips with great motion, most of it came from prompts that describe the shot like a director: lens, movement, subject action, lighting. api_config overrides the endpoint and key per request; skip_error outputs a placeholder instead of aborting the workflow.
Inputs and outputs
prompt(required, multiline, up to 2000 chars).resolution(required, default768P) andduration(required, default 6).- Outputs:
video(VIDEO),request_id,urls.
Installing BizyAirPlus
ComfyUI Manager: "BizyAirPlus", or:
cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt
Restart, account at bizyair.ai, API key, BizyAirPlus button ON. Dependencies: bizyair-cloudberry, bizytrd.
Common issues
The resolution/duration coupling is the main gotcha - pick 1080p and the 10-second option silently isn't available, which confuses the first few runs. Everything else is the pack's standard: missing button means deps in the wrong environment (pip show bizyair-cloudberry), cloud errors live in the console, and you're paying per generation. Fast variants are cheaper than the standard tier, which is the quiet argument for trying this node first.
Inputs (5)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | Video generation prompt words, maximum 2000 characters; support conch mirror instruction writing. | |
| resolution | COMBO | 768P | Supports 768P, 1080p; only 768P is supported when the duration is 10 seconds. |
| duration | COMBO | 6 | 6 seconds, 10 seconds; 1080P only supports 6 seconds. |
| 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 | — |