Runway Gen-4 Turbo
Frames in, motion out, in ComfyUI
- frameImages
- video
Runway's Gen-4 Turbo is the fast lane of the video model that made "consistency" its whole pitch, and this node is it as a ComfyUI drop-in. Feed it one or more frames (the required frameImages input is a plain IMAGE tensor - wire in Load Image or anything that produces frames) and it animates them into a short video. That's the classic Gen-4 trick: lock the character and look in a still, then get motion that keeps them.
The prompt is optional here - positivePrompt defaults to empty. In practice you use it to steer the action ("the person waves, camera slowly pushes in") rather than to describe the scene, because the scene comes from the frames. This is the "I have a great still, give me a living version" node, and the Turbo suffix means it's the fast, cheaper lane of Runway's video stack - drafts, test animations, storyboards.
How it works
frameImages go to runway:1@1 with taskType: videoInference; the returned video is saved to your output folder and wrapped as a native video (VIDEO) output (or a file path on older ComfyUI).
Inputs that matter:
frameImages- required. One or more starting frames.positivePrompt- optional motion/action direction.width/height- 1024×1024 default.seed- reproducible runs.duration- 2–10 seconds, default 10. This is the lever on both length and cost.numberResults- 1–4 candidates.providerSettings.runway.contentModeration.publicFigureThreshold-auto/lowfor public-figure filtering.safety.mode-none/fast/full.outputFormat- MP4 / WEBM / MOV.
Output: video (VIDEO).
Installing
Pack-wide, one install: ComfyUI Manager → Custom Nodes Manager → search Runware → install → restart, or:
cd ComfyUI/custom_nodes
git clone https://github.com/Runware/ComfyUI-Runware
pip install -r ComfyUI-Runware/requirements.txt
No model downloads, light deps (runware-sdk, Pillow, soundfile). API key from the Runware dashboard.
Gotchas
Duration is the cost dial - a 10-second run is five times the compute of 2 seconds, and video is already the most expensive thing in the pack, so check the title-bar cost before you let everything run long. Also, this is Gen-4 Turbo: great for iterating on motion, but if the movement needs to be pristine for a final deliverable, the non-Turbo Gen-4 path is the quality lane. And remember the frames carry the identity - the model will happily animate whatever artifacts are in your still, so clean up the source frame first. Consistency is a feature of the model, not a cleanup step that happens later.
Inputs (14)
| Name | Type | Default | Description |
|---|---|---|---|
| frameImages | IMAGE | — | |
| positivePromptopt | STRING | Text prompt describing elements to include in the generated output. | |
| sizeopt | COMBO | 720p (1:1) | Output size. Only one of resolution or dimensions is sent. |
| seedopt | INT | 00–4294967295 | Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range. |
| durationopt | INT | 102–10 | 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). |
| numberResultsopt | INT | 11–4 | Number of results to generate. Each result uses a different seed, producing variations of the same parameters. |
| providerSettings.runway.contentModeration.publicFigureThresholdopt | COMBO | auto | Public figure filtering strictness. |
| safetyopt | BOOLEAN | false | Enable to set safety. Off uses the model's default. |
| safety.checkContentopt | BOOLEAN | false | Enable or disable content safety checking. |
| safety.modeopt | COMBO | fast | Safety checking mode for video generation. |
| ttlopt | BOOLEAN | false | Enable to set ttl. Off uses the model's default. |
| ttl_valueopt | INT | 60 | Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`. |
| outputFormatopt | COMBO | MP4 | File format for the generated video. |
| outputQualityopt | INT | 9520–99 | Compression quality of the output. Higher values preserve quality but increase file size. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |