generate_image
Generate images from text prompts using AI models. Supports multiple quality tiers and customization for illustrations, photos, logos, and visuals.
Instructions
Generate an image from a text prompt — AI image generation supporting Astria, DALL-E 3, and other models. Use this for any image creation request: illustrations, photos, AI art, logos, icons, graphics, or visuals. Accepts a raw prompt and optional size, quality tier, model, and Astria-specific parameters. Quality tiers: basic (fast/cheap), standard (Nano Banana Gemini 2.5), premium (GPT Image 2.0). Use the intent-specific tools (generate_article_hero_image, etc.) when you know the content type.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | Text description of the image to generate. | |
| size | No | Output image dimensions (default: 1024x1024). | |
| quality | No | Quality tier: basic (fast/cheap), standard (Nano Banana Gemini 2.5), premium (GPT Image 2.0) (default: standard). | |
| model | No | Model to use for image generation (default: determined by quality tier). | |
| style | No | Image style, optional (dall-e-3 only). | |
| astria_style | No | Astria style preset (Astria only). | |
| color_grading | No | Color grading preset (Astria only). | |
| film_grain | No | Add film grain effect (Astria only). | |
| face_swap | No | Use face swap with training images (Astria only). | |
| inpaint_faces | No | Inpaint faces for better quality (Astria only, requires super_resolution). | |
| face_correct | No | Apply face correction (Astria only). | |
| scheduler | No | Scheduler algorithm (Astria only). | |
| hires_fix | No | Use high-resolution fix with super_resolution (Astria only). | |
| controlnet | No | ControlNet model (Astria only). | |
| mask_image_url | No | URL of mask image for inpainting (Astria only). | |
| controlnet_conditioning_scale | No | ControlNet conditioning scale 0.0-1.0 (Astria only). | |
| controlnet_txt2img | No | Use ControlNet for txt2img (Astria only). |