Flow nodes
Its inputs, parameters, and operation methods are explained node by node. Nodes are created on the canvas through the "Add Node" menu, or you can drag a connection from the output port of an existing node and select it directly.
In this chapter
Image generation nodePurpose: Generate pictures based on text descriptions or reference materials.Image editing nodePurpose: Process uploaded images or node output. After creating the node, select the crop, color or annotation function.Video generation nodePurpose: Generate video based on text, image, video or audio input.Video editing nodePurpose: Process uploaded videos or node output. After creating the node, select a loop, clip, score, or composition function.Video upscale nodePurpose: make videos sharper and smoother. With FlowWall DetailX, super-resolution and frame interpolation run in a single pass, taking 480p video up to 4K at 60 fps.Text to speech nodePurpose: Convert text into playable speech audio.Audio generation nodeGenerate an entire audio piece from a description. The soundtrack, multi-character dialogue, and environmental sound effects can be generated separately, or they can be written in the same description and generated at once, and the output is a mixed audio track. The difference from the Text to speech node is that Text to speech reads a piece of text, while this node performs the entire sound scene.Music nodePurpose: Generate music based on lyrics, genre and emotional description.LLM nodePurpose: Process text, analyze pictures or videos, and output prompt words, summaries or other text results.Text nodePurpose: Save and reuse long text; pass @ References assets connected to the same target node.Code animation nodeDescribe the animation you want in one sentence. A language model writes the animation code and the node plays it right away — UI motion, Vlog text, Cinematic title, Cartoon short and more, up to 60 seconds long. When you like it, export it as a video and feed it into Video composition or any other downstream node.Subtitles nodeRecognize lines in video or audio, generate animated Subtitles with timeline, or directly import ready-made Subtitles files. Subtitles will burn into the slice when connected to a compositing node.Grid split nodeCut a picture into grids, and the exported slices will be directly converted into a Grid group, with the frame and dividing lines consistent with the source picture. It is suitable for splitting multi-frame layout diagrams and multi-view characters into materials that can be used independently.