Gemini 3 Pro Image
Gemini 3 Pro Image images model from Google, run through Wavemaker. Pick it in the model picker or name it in your brief. This page gathers its configurations and example graphs.
Discover · Wavemaker catalog
When to use it
- Gemini 3 Pro Image images model from Google, run through Wavemaker. Pick it in the model picker or name it in your brief. This page gathers its configurations and example graphs.
What you provide
- A brief and the image, video or audio references needed for the selected task.
- prompt
- image: attach-scene.png
- image: attach-person.png
What you get
- An explicit gemini model configuration and example graph.
- An editable Comfy graph with the stages listed in Technical details. Generated outputs depend on the selected runtime and bound inputs.
- Output stages: Image.
- Delivers 1 composite still of 4 panels. One file, not separate images. A 2×2 grid of the same product in four lifestyle settings.
Creative guidance
- gemini pro image: gemini model configuration for text-to-image.
- Expand ad frame: Outpaint / expand an existing ad frame to a new aspect while preserving subject, brand props, and lighting continuity.
- Multi-grid product ad: One composite image: a 2×2 product ad grid with consistent lighting across the four cells, for catalog or social carousel frames. The cells are panels in one file, not separate images.
- Product into AI influencer UGC: Places your product into an AI-influencer UGC scene — a natural handheld look with the product held or in frame. Useful for testing creator-style creative without a shoot.
- Insert product into existing ad: Composites your product into an existing ad scene, matching light and perspective so the insert reads as photographed, not pasted. Provide the scene brief and product description.
- Reframe / recompose ad: Recomposes an ad frame for a new placement — reframing the subject and rebalancing negative space instead of blind cropping. Use it to adapt one master visual across aspect ratios.
- Ad scene pair (first + next): Generate a paired first-scene / next-scene still for ad-film continuity — two beats, same cast and product language.
- Fashion: Concept, lookbook, on-model, and try-on workflows.
- Best recipe ports: High-demand commercial shapes: product insert, multi-angle, jewelry, try-on, UGC video, talking avatar, beauty and influencer ads.
- Gear · High quality: Legacy system:high-quality — 2K images, pro video tiers, writingModel=fable companions.
- Mode · Refine: Legacy refine treatments (recast / outfit / face). Bind reference images in the editor.
- Model · Gemini Image: Gemini image generation — prompt-faithful stills with strong text.
- Portrait: Headshot, face detailer, and identity sheet workflows.
- Product: Packshot, lifestyle, background swap, and 360 product stills.
- Writing · Claude Fable: Legacy writingModel=fable (pinned by High quality gear). Companion HQ media set.
- Writing · Claude Opus: Legacy writingModel=opus (claude-opus-4-8). High-craft media companions.
- Design mockup: Place flat artwork onto a realistic product mockup — apparel, packaging, or device — with correct perspective and lighting.
- Clothing & accessories replace: Replace garments or accessories on a model while preserving body pose, face identity, and scene lighting.
- Outfit to editorial shoot: Turns an outfit description into a full editorial shoot look — model, styling, set, and lighting composed like a magazine spread.
- Influencer clothes + pose transfer: Transfer outfit + pose onto a consistent influencer identity for lookbook / UGC stills.
- Identity scene insert: Place a person into a scene. Upload the scene first, then the person — output matches the scene frame.
- Identity try-on: Put a garment on a person. Upload the person first, then the garment — output matches the person frame.
- Style transfer: Apply the style of a reference image to your image while keeping composition and structure locked.
- Gemini Image (2K): Gemini image generation at 2K — the same prompt fidelity with more resolution headroom for crops and print-adjacent uses.
- Gemini Image 4K: Gemini image generation at 4K — maximum detail for hero assets, large-format crops, and zoom-heavy layouts.
- Gemini Omni video: Gemini Omni video generation — prompt-to-video (or image-to-video) through Google's Interactions API, with aspect and duration steered from your brief.
- Gemini Omni edit: A source graph for editing a supplied video through the Gemini Omni video node.
- Gemini Omni extend: A source graph for requesting a continuation of a supplied video through the Gemini Omni video node.
- Gemini Omni image-to-video: Omni image-to-video via existing image widget.
- Gemini Omni reference-to-video: A source graph for generating video through Gemini Omni with image references to guide the result.
- Restore old photo: Restore a damaged or faded photo — scratches, dust, and fading repaired, faces recovered, original identity kept.
- Consistent character still: Lock a character identity into a new scene/wardrobe while keeping face geometry consistent for series work.
- Portrait expression tune: Retarget facial expression on a portrait (smile, soft focus, intensity) while holding identity and head pose.
- Portrait light migration: Portrait relight — migrates a portrait into new lighting (golden hour, studio key, window light) while keeping the person recognizably themselves.
- Jewelry environment scene: Composite jewelry into a premium environment (marble vanity, boutique display) with matched light and reflections.
- Multi-angle product contact sheet: One composite image: a four-panel contact sheet of a product's front, three-quarter, side and macro-detail views with a locked SKU identity. The views are panels in one file, not separate images.
- Product / props replacement: Surgically replace a prop or product inside an existing scene while matching perspective, shadows, and materials.
- Product shot restyle: Restyle an existing product photo into a new studio look (backdrop, lighting, grade) without changing the SKU.
- Studio relight for product: Studio relight — moves your product into a controlled studio lighting setup, fixing flat or mixed-light source photos.
- Refine · Face swap: Refine pass: face swap — replaces the face in an existing image with a reference identity while preserving the original scene and body.
- Refine · Outfit swap: Refine pass: outfit swap — changes what the subject is wearing in an existing image while keeping pose, identity, and scene.
- Refine · Recast: Refine pass: recast — swaps the person in an existing image for a different subject (attach a reference via graph edit) while preserving composition.
- Sketch to render: Turn a sketch or lineart into a finished photoreal render that honors the drawn composition.
- Title card: Full-screen TITLE CARD, HOOK CARD, or INTERTITLE — letters painted into a 2K still, then a locked-camera hold clip.
Setup and requirements
- Name the option you want in your brief; the agent reads its method or configuration (Source entry points) and applies it.
Source licenses and model terms continue to apply. Check runtime availability before generation.
Download source · Learn how to reuse resources