Image Generator 2 works from Blender's 3D Viewport N-panel to convert grey clay renders into photoreal product shots, concept directions, masked edits, and consistent multi-angle image sets. Version 2.1 uses a material-first prompt engine that recognizes the product, assigns PBR-style materials to each visible part, then adds realistic studio lighting, treating grey clay as unpainted material rather than a final color.
Two generation modes are available. Edit Render uses the current Render Result as the source for AI editing, with preservation controls that can guide the provider to keep camera, lighting, shape, material, logos, and label text. Text to Image produces a fresh image from prompt text alone. Keep Maintain Consistency feeds a previous result as reference when you move to a new camera angle, while IMAGINE tracks a render signature covering camera, transform, lens, and resolution and refreshes automatically when any of these change.
Results can appear in a split viewport showing the generated image in an Image Editor, and versions can be upscaled from History or edited within a masked region. The add-on requires Blender 5.0 or newer plus either a Google Gemini or OpenAI API key; provider usage costs are paid directly to the selected provider.