Node reference
Every node you can place in Studio: 117 in total, made up of 8 inputs, 68 AI models, 38 utilities and 3 organization nodes. Each page lists the node's ports, its settings and where it runs.
In Studio, press Tab and type a name to place any of these.
Inputs (8)
Where prompts, media and values enter the graph.
- 3D UploadUpload or drag-drop a .glb model
- Audio UploadUpload or drag-drop an audio file
- Color PickerVisual color with hex output
- Image UploadUpload or drag-drop an image
- NumberNumeric value with optional range
- PromptEnter a prompt or text value
- SeedLock or randomize for reproducibility
- Video UploadUpload or drag-drop a video
AI generation (68)
One node per model. These run on our servers.
- AMT Frame Interpolation10-30s
- Background Removal30-120s
- BiRefNet V22-8s
- Canny Edge3-10s
- Clarity Upscaler5-25s
- Cleanup (Erase)10-30s
- ControlNet (Flux)10-30s
- Depth Map3-10s
- Face Swap (Image)3-10s
- Face Swap (Video)15-60s
- Flux 2 Dev8-20s
- Flux 2 Pro5-15s
- Flux Dev5-15s
- Flux Kontext10-30s
- Flux LoRA5-15s
- Flux Schnell2-6s
- Gemini Omni30-180s
- GPT Image 215-60s
- Grok Imagine 1.530-120s
- Grok Imagine 1.5 I2V30-120s
- HappyHorse 1.0 I2V30-180s
- Hunyuan3D v330-120s
- Hyper3D Rodin30-120s
- ICLight V230-90s
- Ideogram 4.05-20s
- Image Outpaint5-15s
- Imagen 45-15s
- Inpaint (Flux Dev)10-30s
- Inpaint (Flux Pro)10-30s
- Inpaint (Qwen)5-25s
- Kling 3.030-120s
- Kling 3.0 I2V30-120s
- Kling 3.0 Pro60-180s
- Kling 3.0 Pro I2V60-180s
- Kling O1 FLF2V30-90s
- Kling O1 Reference30-120s
- Kling O330-120s
- Kling O3 I2V30-120s
- Kling O3 Pro60-180s
- Kling O3 Reference30-120s
- Line Art3-10s
- MMAudio V25-15s
- Nano Banana 23-15s
- Nano Banana Pro5-25s
- Object Removal5-15s
- Qwen Edit Plus10-30s
- Qwen Image 2.05-20s
- Qwen Image Edit5-20s
- Recraft V45-15s
- SAM 3 Image5-30s
- SAM 3 Video30-120s
- Seedance 2.030-180s
- Seedance 2.0 I2V30-180s
- Seedance 2.0 Reference240-540s
- SHARP Gaussian10-60s
- Text to Speech3-15s
- Topaz Image5-20s
- Topaz Video60-300s
- Trellis10-60s
- Video BG Removal15-60s
- Vidu Q1 FLF2V15-60s
- WAN 2.730-120s
- WAN 2.7 I2V30-120s
- WAN 2.7 Image5-20s
- WAN 2.7 Image Pro10-40s
- WAN Effects30-90s
- WAN Pro I2V30-120s
- WAN V2V20-90s
Utility (38)
Compositing, matte, text and video operations, in your browser.
- 3D CompositePlace 3D model over background
- 3D TurntableRender rotating turntable video
- ArraySplit text by delimiter into items
- Blur / SoftenGaussian blur with adjustable radius
- Channel SplitterSplit image into R, G, B grayscale channels
- Crop / TransformCrop, resize, rotate, flip
- DifferenceShow pixel changes between A and B
- Frame ExtractorExtract a still frame from video
- GradeLift / Gamma / Gain + BCSH with mask support
- Image IteratorOutput URLs one at a time (comma-separated input)
- InvertInvert colors or masks
- ListSelect one item from a list
- Mask ExtractorExtract transparency as grayscale mask (use SAM 3 for JPEGs)
- Mask PainterPaint a mask by hand
- Matte EdgeDetect edges in alpha channel
- Matte Grow / ShrinkDilate or erode alpha channel
- Merge / Composite16 blend modes
- Merge AlphaCombine RGB image with alpha mask
- Motion AnnotatorPaint motion cues on an image
- PreviewDisplay result in clean view
- Prompt EnhancerAI prompt improvement
- ReformatConform an image or video to a delivery format
- ResizeQuick image scale (Fit keeps aspect, no bars)
- SharpenUnsharp mask sharpening
- Splat CinematographerAuthor a camera path through a splat & render to video
- Splat CropBox-crop a gaussian splat
- Splat ViewerView gaussian splat & capture a frame
- SplitterFan-out one input to multiple outputs
- Text CaseConvert text to upper/lower/title case
- Text ConcatenatorJoin text inputs
- Text IteratorOutput lines one at a time across runs
- Text ReplaceFind and replace text
- Text SplitSplit text by delimiter and pick a part
- Text TemplateInsert text into a template with {input} placeholder
- Video DifferencePer-frame |A - B| video diff
- Video IteratorOutput video URLs one at a time
- Video MergeComposite two clips with matte + blend modes
- Vision CaptionerAI describes an image as text
Organization (3)
Visual structure. No processing.