ComfyUI Node

HappyHorse 1.1

Alibaba's sleeper video model with the size presets

By Runware·Created 2 years ago·Updated about a month ago· 140
HappyHorse 1.1
  • referenceImages
  • frameImages
  • video
positivePrompt
size720p (16:9)
seed0
duration5
numberResults1
safetyfalse
safety.checkContentfalse
safety.modefast
ttlfalse
ttl_value60
outputFormatMP4
outputQuality95

HappyHorse 1.1 is the Alibaba video generation entry in the Runware pack, and it's the rare node here where the UI tells you exactly how the vendor thinks about video: there's a size dropdown with 13 presets, from "None (model default)" through 720p and 1080p in every aspect ratio (16:9, 9:16, 1:1, ~4:3, ~3:4) to "from input" variants. No fiddly pixel math - you pick your delivery format and go. That's HappyHorse's whole vibe: a practical, output-format-first video model rather than a raw-resolution playground.

It's a hybrid input node. You can feed referenceImages for style or subject reference, frameImages for image-to-video (or multi-frame), and positivePrompt on top - or any combination. In practice most people land on: still from an image model in frameImages, a motion prompt, a size preset, and a duration. It's also the rare video node in the pack that gives you a proper duration slider as a real number (3 to 15 seconds, default 5) instead of a toggle - which says something about how the model is meant to be driven.

What you set

  • positivePrompt - what happens in the clip.
  • frameImages / referenceImages - the visual anchors.
  • size - the preset dropdown. The tooltip notes only one of resolution or dimensions is sent, so pick a preset and don't also try to set raw dims (there aren't any - this node is preset-driven).
  • duration - 3 to 15 seconds, step 1, default 5.
  • seed - up to 2^31-1, standard.
  • numberResults 1-4, safety.mode none/fast/full, outputFormat MP4/WEBM/MOV, outputQuality 20-99.

Output is video (VIDEO).

How it works

Standard pack machinery - model AIR + taskType over REST via the Runware SDK, media converted back to a native VIDEO, cost on the title bar. Video is the most expensive modality in the pack per run, and HappyHorse is no exception, but the preset system keeps you honest: pick 720p for drafts, promote to 1080p for finals. The duration slider is a real saving, too - a 3-second test clip costs a fraction of a 15-second one, and you don't need the long cut to test whether the motion works.

Installing

ComfyUI Manager → search Runware → install → restart. Manual:

cd ComfyUI/custom_nodes
git clone https://github.com/Runware/ComfyUI-Runware
pip install -r ComfyUI-Runware/requirements.txt

No model downloads; dependencies are runware-sdk, pillow, soundfile. API key via Settings → Runware API key, RUNWARE_API_KEY, or runware auth login.

Troubleshooting

Key first, always. If your video comes back at the wrong aspect, check the size preset rather than blaming the input - the preset wins. And if you're coming from the Hailuo nodes expecting a width/height pair, there isn't one here; that's the HappyHorse design, and the presets cover everything you'd set anyway.

CategoryRunware/Video/alibaba

Inputs (14)

NameTypeDefaultDescription
referenceImagesoptIMAGE
frameImagesoptIMAGE
positivePromptoptSTRINGText prompt describing elements to include in the generated output.
sizeoptCOMBO720p (16:9)Output size. Only one of resolution or dimensions is sent.
seedoptINT00–2147483647Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.
durationoptINT53–15Length of the generated video in seconds. The total number of frames produced is determined by duration multiplied by the model's frame rate (fps).
numberResultsoptINT11–4Number of results to generate. Each result uses a different seed, producing variations of the same parameters.
safetyoptBOOLEANfalseEnable to set safety. Off uses the model's default.
safety.checkContentoptBOOLEANfalseEnable or disable content safety checking.
safety.modeoptCOMBOfastSafety checking mode for video generation.
ttloptBOOLEANfalseEnable to set ttl. Off uses the model's default.
ttl_valueoptINT60Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.
outputFormatoptCOMBOMP4File format for the generated video.
outputQualityoptINT9520–99Compression quality of the output. Higher values preserve quality but increase file size.

Outputs (1)

NameTypeDescription
videoVIDEO