diff --git a/scientific-skills/generate-image/SKILL.md b/scientific-skills/generate-image/SKILL.md index 25f6aaf..8ed8e3d 100644 --- a/scientific-skills/generate-image/SKILL.md +++ b/scientific-skills/generate-image/SKILL.md @@ -1,6 +1,6 @@ --- name: generate-image -description: Generate or edit images using AI models (FLUX, Gemini). Use for general-purpose image generation including photos, illustrations, artwork, visual assets, concept art, and any image that is not a technical diagram or schematic. For flowcharts, circuits, pathways, and technical diagrams, use the scientific-schematics skill instead. +description: Generate or edit images using AI models (FLUX, Nano Banana 2). Use for general-purpose image generation including photos, illustrations, artwork, visual assets, concept art, and any image that is not a technical diagram or schematic. For flowcharts, circuits, pathways, and technical diagrams, use the scientific-schematics skill instead. license: MIT license compatibility: Requires an OpenRouter API key metadata: @@ -9,7 +9,7 @@ metadata: # Generate Image -Generate and edit high-quality images using OpenRouter's image generation models including FLUX.2 Pro and Gemini 3 Pro. +Generate and edit high-quality images using OpenRouter's image generation models including FLUX.2 Pro and Gemini 3.1 Flash Image Preview. ## When to Use This Skill @@ -58,18 +58,18 @@ The script will automatically detect the `.env` file and provide clear error mes ## Model Selection -**Default model**: `google/gemini-3-pro-image-preview` (high quality, recommended) +**Default model**: `google/gemini-3.1-flash-image-preview` (high quality, recommended) **Available models for generation and editing**: -- `google/gemini-3-pro-image-preview` - High quality, supports generation + editing +- `google/gemini-3.1-flash-image-preview` - High quality, supports generation + editing - `black-forest-labs/flux.2-pro` - Fast, high quality, supports generation + editing **Generation only**: - `black-forest-labs/flux.2-flex` - Fast and cheap, but not as high quality as pro Select based on: -- **Quality**: Use gemini-3-pro or flux.2-pro -- **Editing**: Use gemini-3-pro or flux.2-pro (both support image editing) +- **Quality**: Use gemini-3.1-flash-image-preview or flux.2-pro +- **Editing**: Use gemini-3.1-flash-image-preview or flux.2-pro (both support image editing) - **Cost**: Use flux.2-flex for generation only ## Common Usage Patterns @@ -115,7 +115,7 @@ python scripts/generate_image.py "Image 2 description" --output image2.png - `prompt` (required): Text description of the image to generate, or editing instructions - `--input` or `-i`: Input image path for editing (enables edit mode) -- `--model` or `-m`: OpenRouter model ID (default: google/gemini-3-pro-image-preview) +- `--model` or `-m`: OpenRouter model ID (default: google/gemini-3.1-flash-image-preview) - `--output` or `-o`: Output file path (default: generated_image.png) - `--api-key`: OpenRouter API key (overrides .env file) @@ -172,7 +172,7 @@ If the script fails, read the error message and address the issue before retryin - Be specific about what changes you want (e.g., "change the sky to sunset colors" vs "edit the sky") - Reference specific elements in the image when possible - For best results, use clear and detailed editing instructions -- Both Gemini 3 Pro and FLUX.2 Pro support image editing through OpenRouter +- Both Gemini 3.1 Flash Image Preview and FLUX.2 Pro support image editing through OpenRouter ## Integration with Other Skills diff --git a/scientific-skills/generate-image/scripts/generate_image.py b/scientific-skills/generate-image/scripts/generate_image.py index 1dd71d1..9c613b8 100755 --- a/scientific-skills/generate-image/scripts/generate_image.py +++ b/scientific-skills/generate-image/scripts/generate_image.py @@ -3,7 +3,7 @@ Generate and edit images using OpenRouter API with various image generation models. Supports models like: -- google/gemini-3-pro-image-preview (generation and editing) +- google/gemini-3.1-flash-image-preview (generation and editing) - black-forest-labs/flux.2-pro (generation and editing) - black-forest-labs/flux.2-flex (generation) - And more image generation models available on OpenRouter @@ -74,7 +74,7 @@ def save_base64_image(base64_data: str, output_path: str) -> None: def generate_image( prompt: str, - model: str = "google/gemini-3-pro-image-preview", + model: str = "google/gemini-3.1-flash-image-preview", output_path: str = "generated_image.png", api_key: Optional[str] = None, input_image: Optional[str] = None @@ -84,7 +84,7 @@ def generate_image( Args: prompt: Text description of the image to generate, or editing instructions - model: OpenRouter model ID (default: google/gemini-3-pro-image-preview) + model: OpenRouter model ID (default: google/gemini-3.1-flash-image-preview) output_path: Path to save the generated image api_key: OpenRouter API key (will check .env if not provided) input_image: Path to an input image for editing (optional) @@ -212,7 +212,7 @@ def main(): formatter_class=argparse.RawDescriptionHelpFormatter, epilog=""" Examples: - # Generate with default model (Gemini 3 Pro Image Preview) + # Generate with default model (Gemini 3.1 Flash Image Preview) python generate_image.py "A beautiful sunset over mountains" # Use a specific model @@ -228,7 +228,7 @@ Examples: python generate_image.py "Add a hat to the person" --input portrait.png -m "black-forest-labs/flux.2-pro" Popular image models: - - google/gemini-3-pro-image-preview (default, high quality, generation + editing) + - google/gemini-3.1-flash-image-preview (default, high quality, generation + editing) - black-forest-labs/flux.2-pro (fast, high quality, generation + editing) - black-forest-labs/flux.2-flex (development version) """ @@ -243,8 +243,8 @@ Popular image models: parser.add_argument( "--model", "-m", type=str, - default="google/gemini-3-pro-image-preview", - help="OpenRouter model ID (default: google/gemini-3-pro-image-preview)" + default="google/gemini-3.1-flash-image-preview", + help="OpenRouter model ID (default: google/gemini-3.1-flash-image-preview)" ) parser.add_argument(