# Edit nodes Edit nodes change something that already exists: a prompt, an image or a video. Several are free. Last updated: 2026-10-05 **Key facts** - Spacelab edit nodes change something that already exists, a prompt, an image or a video, and Crop, Rotate & Flip, Colour Grade, Image Resize, Extract Frame and Trim video are free. - The Spacelab Upscaler enlarges an image 2x, 4x or 6x with Topaz and outputs JPG or PNG. - Spacelab's Image Resize scales an image to exact pixels, up to 7680 pixels on a side, with Fill, Fit or Stretch. - Spacelab's Video edit node changes or extends a video from a prompt with Gemini Omni or Seedance, and Gemini Omni 1.1 Flash extends in 10-second steps up to 40 seconds in total. Web page: https://sloosh.ai/docs/nodes-edit Each entry says what the node does, then gives the same lines in the same order: Best for, In, Out, Settings, Cost, Limits, Example and Tip. A line that doesn't apply is left out. To find a node on the canvas, search Add node by name. ## At a glance - [Prompt Enhancer](https://sloosh.ai/docs/nodes-edit.md#prompt-enhancer): turns a rough idea into a detailed prompt. - [BG Remove](https://sloosh.ai/docs/nodes-edit.md#bg-remove): cuts the subject out onto a transparent background. - [Auto Mask](https://sloosh.ai/docs/nodes-edit.md#auto-mask): selects what you name, as a cut-out and a mask. - [Crop](https://sloosh.ai/docs/nodes-edit.md#crop): cuts an image to a size in pixels. Free. - [Rotate & Flip](https://sloosh.ai/docs/nodes-edit.md#rotate-flip): turns or mirrors an image or video. Free. - [Colour Grade](https://sloosh.ai/docs/nodes-edit.md#colour-grade): fixes colour on an image or video. Free. - [Upscaler](https://sloosh.ai/docs/nodes-edit.md#upscaler): enlarges an image 2x, 4x or 6x. - [Image Resize](https://sloosh.ai/docs/nodes-edit.md#image-resize): scales an image to exact pixels. Free. - [Video edit](https://sloosh.ai/docs/nodes-edit.md#video-edit): changes or extends a video from a prompt. - [Extract Frame](https://sloosh.ai/docs/nodes-edit.md#extract-frame): pulls one frame out of a video. Free. - [Video Concat](https://sloosh.ai/docs/nodes-edit.md#video-concat): joins clips end to end. - [Veed Subtitles](https://sloosh.ai/docs/nodes-edit.md#veed-subtitles): burns subtitles into a video. - [Merge audio + video](https://sloosh.ai/docs/nodes-edit.md#merge-audio-video): puts an audio track on a video. ## Prompts ### Prompt Enhancer Node: Prompt Enhancer (`prompt-enhancer`) - Inputs: `prompt` (Prompt, text), `reference_images__0` (Image 1, image), `documents__0` (Document 1, document) - Outputs: `text` (Enhanced Prompt, text) Turns a short description into a detailed prompt. Pick a prompt type, such as Enhance Image Prompt, Craft Video Prompt or Static Ads Prompter, and it fills in the model's instructions. You can edit them. - Best for: the step between a rough prompt and an image or video model. For anything else a text model does, use Any LLM. - In: Prompt (required) and any number of images. - Out: Enhanced Prompt. - Cost: credits by length of the prompt and answer, by model. - Example: Prompt (red sneaker, rooftop, sunset) → Prompt Enhancer (Enhance Image Prompt) → Image Model. ## Images ### BG Remove Node: BG Remove (`bg-remove`) - Inputs: `image` (Image, image) - Outputs: `image` (Result, image) Cuts the subject out and returns a PNG with a transparent background. - In: Image. - Out: Image. - Cost: credits per image. - Example: product photo → BG Remove → Image Model reference, so the model gets the product without the old background. ### Auto Mask Node: Auto Mask (`auto-mask`) - Inputs: `image` (Image, image), `prompt` (What to mask, setting) - Outputs: `image` (Cut-out, image), `mask` (Mask, image) Selects the thing you name, such as the bag or the shoes, and returns a cut-out that keeps only that thing on a transparent background, plus a black-and-white mask (white is selected). Every match in the image is selected. - Best for: pulling one item out of a busy photo. To keep the whole subject, use BG Remove. - In: Image, and What to mask (type it, or wire a text node). - Out: Cut-out and Mask, both PNG at the image's own size. A space shows the Cut-out. - Settings: What to mask, and Invert to select everything else. - Cost: credits per image. - Limits: images up to 25 MB and 40 megapixels. The run fails when nothing in the image matches. - Example: product photo → Auto Mask (the handbag) → Cut-out into an Image Model reference. - Tip: name one thing in a few plain words. "The red handbag" works better than a sentence. ### Crop Node: Crop (`image-crop`) - Inputs: `image` (Image, image) - Outputs: `image` (Result, image) Crops to a width and height in pixels, centred unless you set X and Y offsets. - In: Image. - Out: Image. - Settings: Aspect ratio, Width, Height, X and Y. The aspect ratio control helps you pick a size but doesn't crop by itself. - Cost: Free. ### Rotate & Flip Node: Rotate & Flip (`rotate-flip`) - Inputs: `media` (Media, media) - Outputs: `media` (Media, media) Rotates an image or video by 90, 180 or 270 degrees clockwise, and flips it horizontally or vertically. - In: an image or a video. - Out: the same type as the input. - Cost: Free. ### Colour Grade Node: Colour Grade (`color-grade`) - Inputs: `image` (Image, image), `video` (Video, video) - Outputs: `image` (Image, image), `video` (Video, video) Adjusts exposure, contrast, highlights, shadows, temperature, tint, saturation and hue, on an image or a video. Set Target colour to change one colour and leave the rest alone. - Best for: small colour fixes, instead of regenerating. - In: Image or Video (wire one). - Out: the matching one. - Cost: Free. - Example: to turn a red jacket blue without touching the model, set Target colour to the red and move Hue until the jacket turns blue. ### Upscaler Node: Upscaler (`upscaler`) - Inputs: `image` (Image, image) - Outputs: `image` (Result, image) Enlarges an image 2x, 4x or 6x with Topaz. - Best for: the last step, after your other edits. - In: Image. - Out: Image. - Settings: Upscale factor (2x, 4x or 6x), Enhance model (Standard V2, Low Resolution V2, CGI, High Fidelity V2 or Text Refine), Output format (JPG or PNG), Subject detection, and Face enhancement for portraits. - Cost: credits per image. ### Image Resize Node: Image Resize (`image-resize`) - Inputs: `image` (Image, image) - Outputs: `image` (Result, image) Resizes an image to exact pixels. Fill crops to fit, Fit adds bars, Stretch distorts. - Best for: hitting a delivery size such as 1080 by 1350. To cut part of the image away, use Crop. - In: Image. - Out: Image. - Settings: width and height, which start at the wired image's own size (or 1920 by 1080), and Fill, Fit or Stretch. - Cost: Free. - Limits: 7680 pixels on a side. ## Video ### Video edit Node: Video edit (`video-edit`) - Inputs: `video` (Video, video), `prompt` (Edit prompt, text), `reference_images__0` (Reference image 1, image), `reference_audios__0` (Reference audio 1, audio) - Outputs: `video` (Result, video) Changes an existing video from a prompt: restyle it, add or remove something, change the light. Some models can also extend a clip. - In: Video (required), Edit prompt (required). - Out: Video. - Settings: Model (Gemini Omni or Seedance), Task (Edit or Extend, where the model has it), Extend by, Resolution. - Cost: credits per second of video, by model and resolution. The node's Run button shows the price for the model and settings you picked. - Limits: Gemini Omni takes one source video of up to 10 seconds, in 16:9 or 9:16. Gemini Omni 1.1 Flash extends in 10-second steps, up to 40 seconds in total. Seedance 2.5 needs at least 4 seconds of source video to edit. - Example: Video Input → Video edit (make it look like a 1970s film, keep everything else) → Video Concat. ### Extract Frame Node: Extract Frame (`video-frame-extract`) - Inputs: `video` (Video, video) - Outputs: `image` (Frame image, image) Pulls one frame out of a video as a JPEG. - Best for: taking a clip's last frame as the next clip's first frame, so two shots join up. - In: Video. - Out: Image. - Settings: Frame, the frame number counting from 0. FPS (25 by default) only labels the frame's timecode. - Cost: Free. - Limits: a Frame past the end of the clip fails the run. ### Trim video Node: Trim video (`video-trim`) - Inputs: `video` (Video, video), `start` (Start (s), setting), `end` (End (s), setting) - Outputs: `video` (Trimmed video, video) Keeps the part of a video between Start and End, in seconds. Decimals work (1.5). Leave End empty to keep everything after Start; an End past the clip keeps it to the end. A Start at or past the end of the clip fails the run, and one trim keeps at most 20 minutes (at 1080p; less above it) of a file up to 2 GB. The cut keeps the clip's frame rate and sound. Free. - In: Video, optional Start and End. Out: Video. - Wire a Number node into Start or End to let a space's visitor choose the cut. ### Video Concat Node: Video Concat (`video-concat`) - Inputs: `videos__0` (Clip 1, video) - Outputs: `video` (Result, video) Joins clips end to end, in the order of its Clip ports. Free. - In: 2 to 10 clips. - Out: Video. - Settings: Mute audio drops the sound. - Cost: free. ### Veed Subtitles Node: Veed Subtitles (`video-subtitle`) - Inputs: `video` (Video, video) - Outputs: `video` (Video with subtitles, video) Transcribes the speech and burns styled subtitles into the video. - In: Video. - Out: Video with subtitles. - Settings: preset, language, position and shadow, plus fonts and colours if the preset isn't right. - Cost: credits per run. ### Merge audio + video Node: Merge audio + video (`merge-audio-video`) - Inputs: `video` (Video, video), `audio` (Audio, audio) - Outputs: `video` (Video, video) Puts an audio track on a video. The video's own sound is replaced. Free. - In: Video, Audio. - Out: Video. - Cost: free. - Limits: the result is always as long as the video. Longer audio is cut off, shorter audio ends in silence. - Example: Video → Merge audio + video ← Lyria 3. ## Frequently asked questions ### Which Spacelab node removes a background? BG Remove cuts the subject out of an image in Spacelab and returns a PNG with a transparent background. To pull one named item out of a busy photo, such as the handbag, use Auto Mask instead, which also returns a black-and-white mask. ### How do I change the colour of one item in Spacelab without regenerating? Use Spacelab's free Colour Grade node: set Target colour to the colour you want to change, then move Hue. For example, set Target colour to a red jacket's red and move Hue until the jacket turns blue, leaving the model untouched. ### How do I join two AI video clips in Spacelab? Wire the clips into Spacelab's Video Concat node, which joins 2 to 10 clips end to end in the order of its Clip ports. To make the second shot start where the first ends, use Extract Frame to take the first clip's last frame as the second clip's First frame. ### How do I resize an image to 1080 by 1350 in Spacelab? Use Spacelab's free Image Resize node and set the width and height. Fill crops to fit, Fit adds bars and Stretch distorts. To cut part of the image away rather than scale it, use the Crop node. ### Can Spacelab add subtitles to a video? Yes. Spacelab's Veed Subtitles node transcribes the speech in a video and burns styled subtitles into it. You choose the style, language, position and shadow, plus fonts and colours. Veed Subtitles costs credits per run. ### How long a source video can Spacelab's Video edit take? Gemini Omni in Spacelab's Video edit takes one source video of up to 10 seconds, in 16:9 or 9:16. Seedance 2.5 needs at least 4 seconds of source video to edit. Both change a clip from an edit prompt, such as a restyle or new light.