AI Image Generation Tool (Bulk Supported)

Software Usage Instructions

1. Configure API settings: Switch to the "API Settings" tab, select an API provider (supports OpenAI, Google Gemini, Stability AI, etc.), and enter the API domain, model name, and API key. Click "Test Connection" to verify that the configuration is correct. After confirming, click "Save Settings"; the configuration will be automatically restored on your next visit. The API key is saved only in your local browser and will not be uploaded to the server.
2. Select a processing mode:
   • Image generation: Suitable for generating a small number of images at a time, supporting three modes: text-to-image, image-to-image, and multi-image fusion
   • Batch processing: Suitable for generating a large number of images at once or applying unified processing to multiple images
3. Select a generation type:
   • Text-to-image: Generate a completely new image from text descriptions only, with no reference image required
   • Image-to-image: Upload a reference image, and the AI edits or transforms its style based on the prompt
   • Multi-image fusion: Upload 2-6 images, and the AI combines content from multiple images to generate a new fused image
4. Select style presets: Choose from 10 style presets, such as realistic photography, anime illustration, oil painting, watercolor, pixel art, 3D rendering, cyberpunk, and Chinese ink painting. After clicking, they are automatically appended to the prompt, and multiple selections can be combined.
5. Enter prompts: Describe in detail the image effect you want to generate, or click a prompt template (cute animals, natural scenery, portraits, product display, sci-fi scenes, food photography) to load one quickly. You can also enter negative prompts to tell the AI what not to generate (such as blur, low quality, watermarks, etc.) to improve generation quality.
6. Configure generation settings:
   • Number of images: Set the number of images to generate (up to 5 per single generation, up to 100 in batch)
   • Image size: Supports multiple ratios, including auto, 1024×1024 square, 1536×1024 landscape, 1024×1536 portrait, 1792×1024 widescreen, and more
   • Output format: Supports three formats: PNG (lossless), JPEG (smaller), and WebP (efficient)
   • Output quality: Selectable from 70% to 95%; the higher the quality, the larger the file
   • Creativity: Controls the randomness of generation. Conservative mode is more stable, while creative mode is more diverse
   • Retry attempts: Automatically retries when an API call fails, using an exponential backoff strategy to improve the success rate
   • Concurrency: The number of tasks processed simultaneously during batch processing; 2-3 is recommended
7. Start generation: Click the "Start Generation" or "Start Batch Generation" button to view the progress bar, elapsed time, and processing logs in real time. During batch processing, you can click "Stop Processing" at any time to interrupt the task.
8. View and download results: After generation is complete, you can preview thumbnails in the results area. Click an image to open a large preview (supports switching with the left and right arrow keys), click the download button for a single image, or click "Download All (ZIP)" to download all images as a package.
9. Generation history: Switch to the "Generation History" tab to view previous generation records (including prompts and thumbnails), and click to quickly reuse historical prompts.

FAQ