Nano Banana

Gemini-native text-to-image and image editing with three model tiers, batch generation, and custom gateway support.

- Three model aliases (nanobanana, nanobanana-2, nanobanana-pro) map to distinct Gemini image models with different speed, cost, and fidelity tradeoffs

- Supports single generation, batch variants, multi-image editing with local file references, and dry-run request inspection before sending

- Strict option validation enforces model-specific constraints (e.g., --size only on Gemini 3, 512 resolution only on nanobanana-2)

- Custom Gemini-compatible gateway support via --base-url and flexible auth modes for self-hosted or proxied endpoints

Nano Banana

A single Python entrypoint for Gemini-native Nano Banana image generation and editing, with model aliases, strict option validation, batch runs, and custom endpoint support.

Workflow

- Open references/config.md to choose environment variables and override order.

- Open references/models-and-api.md to pick the right Nano Banana tier and check model-specific constraints.

- Prefer gemini-3.1-flash-image-preview (nanobanana-2) unless you need either the fastest low-cost default (nanobanana) or the highest-fidelity reasoning model (nanobanana-pro).

- Run scripts/nanobanana.py generate for one request or scripts/nanobanana.py batch for repeated variants.

- Add --dry-run first when the main risk is the payload shape, endpoint, or model-specific option support.

- Pass --base-url or GEMINI_BASE_URL when you need a custom Gemini-compatible gateway.

- Add --save-response <path> on generate when you need the raw JSON body for debugging.

Commands

Single text-to-image request: