ComfyUI Node

PixVerse V4.5

Where the TikTok effect templates live

By Runware·Created 2 years ago·Updated about a month ago· 140
PixVerse V4.5
  • frameImages
  • video
positivePrompt
negativePrompt
size720p (16:9)
seed0
duration(default)
numberResults1
providerSettings.pixverse.cameraMovement(default)
providerSettings.pixverse.effect(default)
providerSettings.pixverse.motionModenormal
safetyfalse
safety.checkContentfalse
safety.modefast
providerSettings.pixverse.style(default)
ttlfalse
ttl_value60
outputFormatMP4
outputQuality95

If you've scrolled short-form video in the last couple of years, you've seen PixVerse V4.5's output: those viral "subject 3 fever" and "kiss me ai" clips, the AI characters doing TikTok dances in weirdly consistent video. That's because V4.5 is the generation where PixVerse leaned hard into one-click effect templates - and this node gives you those exact templates as dropdown options inside ComfyUI, running on Runware's cloud.

The framing that matters: V4.5 is not the newest PixVerse (V6 is), but it's the version with the effect catalog baked in, and that's its reason to exist in 2026. If you want "make this person do the skeleton dance," a template is dramatically more reliable than trying to prompt your way there. It's a fun node, and it's also a genuinely practical one for a niche: anyone producing short-form content who wants the machine's signature effects without subscribing to PixVerse's app.

Inputs that matter

positivePrompt required. The PixVerse-specific controls:

  • providerSettings.pixverse.effect - the template dropdown, and the reason this node exists. Choices include jiggle jiggle, hug your love, kiss me ai, subject 3 fever, skeleton dance, the tigers touch, fin-tastic mermaid, boom drop, and more. These are the virality levers, straight from the platform.
  • providerSettings.pixverse.cameraMovement - cinematic camera moves: horizontal_left/right, vertical_up/down, zoom_in/out, crane_up, smooth_zoom_in, plus auto_camera. Pair a template with a camera move and you're directing.
  • providerSettings.pixverse.motionMode - normal or fast. Motion intensity. fast is for the energetic template dances; normal keeps things sane.
  • providerSettings.pixverse.style - anime, 3d_animation, clay, comic, cyberpunk, or default. Same aesthetic presets as the newer PixVerse nodes.
  • frameImages - first-frame image-to-video; drop in a character and the effect templates take over.
  • duration - (default), 5, or 8 seconds.
  • size - the resolution/aspect dropdown from 360p to 1080p, including "from input" variants.

negativePrompt, seed, numberResults (up to 4), outputFormat (MP4/WEBM/MOV), outputQuality, and safety are the standard tail.

How it works

The node sends pixverse:1@3 as a videoInference task through the Runware SDK, with the effect and camera settings embedded in providerSettings. The finished clip returns as a native VIDEO socket for SaveVideo or any video consumer. Cloud-only, pay-per-run - and the title bar's cost line is worth watching, because effect-heavy clips are not the cheapest.

Install and gotchas

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

Restart; API key via Settings → "Runware API key", RUNWARE_API_KEY, or runware auth login.

The honest catch: effect templates are a look, and the look ages fast. V4.5's templates are the current viral set, but "skeleton dance" energy has a shelf life, and you're paying a commercial API for what is fundamentally a meme factory. If your workflow is serious cinematic video, the V6 node is the better tool; if your workflow is short-form volume, this is the one you actually reach for. Just don't confuse "can generate the viral effect" with "original art."

CategoryRunware/Video/pixverse

Inputs (18)

NameTypeDefaultDescription
positivePromptSTRINGText prompt describing elements to include in the generated output.
frameImagesoptIMAGE
negativePromptoptSTRINGPrompt to guide what to exclude from generation. Ignored when guidance is disabled (CFGScale ≤ 1).
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.
durationoptCOMBO(default)Length 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.
providerSettings.pixverse.cameraMovementoptCOMBO(default)Cinematic camera movement.
providerSettings.pixverse.effectoptCOMBO(default)Visual effect template.
providerSettings.pixverse.motionModeoptCOMBOnormalMotion intensity in the generated video.
safetyoptBOOLEANfalseEnable to set safety. Off uses the model's default.
safety.checkContentoptBOOLEANfalseEnable or disable content safety checking.
safety.modeoptCOMBOfastSafety checking mode for video generation.
providerSettings.pixverse.styleoptCOMBO(default)Artistic style aesthetic 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