Gemini Nano Banana 2 - Image Generation & Editing

Generate new images or edit existing ones using Google's Gemini Nano Banana 2 model (gemini-3.1-flash-image-preview) via the Gemini API. This is the latest and most capable Gemini image model, featuring advanced reasoning ("Thinking") with controllable levels, high-fidelity text rendering, Google Web + Image Search grounding, output from 512 up to 4K resolution, and 14 aspect ratios.

Usage

Run the script using absolute path (do NOT cd to skill directory first):

Generate new image:

uv run ~/.claude/skills/gemini-nano-banana-2/scripts/generate_image.py --prompt "your image description" --filename "output.png" [--aspect-ratio 1:1|1:4|1:8|2:3|3:2|3:4|4:1|4:3|4:5|5:4|8:1|9:16|16:9|21:9] [--resolution 512|1K|2K|4K] [--output-format png|webp|jpeg] [--response-modality text-and-image|image-only] [--thinking-level minimal|high] [--google-search] [--image-search] [--yes] [--cost-threshold 0.10] [--api-key KEY]

Edit existing image:

uv run ~/.claude/skills/gemini-nano-banana-2/scripts/generate_image.py --prompt "editing instructions" --filename "output.png" --input-image "path/to/input.png" [--aspect-ratio 1:1] [--resolution 2K] [--api-key KEY]