Image Generation lets the AI create visual content from text descriptions. Use it to produce diagrams, schematics, concept illustrations, and flow visualizations when you need a quick visual reference.
Use cases
Image Generation is useful for creating visuals that help communicate simulation concepts:
- Flow schematics -- Visualize expected flow patterns, recirculation zones, or wake structures before running a simulation
- Geometry diagrams -- Illustrate the simulation domain, boundary conditions, and dimensions
- Concept illustrations -- Show physical phenomena like heat transfer mechanisms, turbulence structures, or phase interfaces
- Presentation graphics -- Generate visuals for reports or presentations that explain your simulation setup
Image Generation creates AI-generated illustrations, not simulation results. For actual flow visualization from simulation data, use the
3D Viewer after running a simulation.
How it works
Describe what you want
Tell the AI what image to create. Be specific about the content, layout, and style. For example: "Draw a schematic of a backward-facing step with flow direction arrows and the recirculation zone labeled."
AI generates the image
The request is sent to the image generation model. Generation typically takes 5-15 seconds.
Review the result
The generated image appears in the chat as a thumbnail. Click to expand it in a full-screen modal for detailed viewing.
Iterate if needed
Ask the AI to adjust the image: "Make the arrows larger" or "Add dimension labels." Each revision generates a new image.
Tips for better results
- Be specific -- "A 2D cross-section diagram of a pipe with a sudden expansion from 5cm to 10cm diameter, showing flow direction and expected separation region" produces better results than "pipe flow diagram."
- Specify the style -- Request "technical schematic," "annotated diagram," or "engineering drawing" to get appropriate visual styles.
- Include labels -- Ask for labeled dimensions, boundary condition annotations, or flow direction indicators.
- Reference known setups -- Mention standard problems like "lid-driven cavity" or "flow over a cylinder" for recognizable layouts.
Configuration
| Parameter | Value |
|---|
| Toggle | Can be enabled or disabled in Settings |
Display
Generated images appear as clickable thumbnails within the chat. Clicking opens an expandable modal with the full-resolution image. You can right-click to save the image locally or ask the AI to include it in an exported report.