Nano Banana 2 Image Generation & Editing

Generate new images or edit existing ones using Google's Nano Banana 2 API (Gemini 3.1 Flash Image Preview).

Usage

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

Generate new image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "your image description" --filename "output.png" [--model MODEL] [--resolution 512|1K|2K|4K] [--aspect-ratio RATIO] [--thinking-level minimal|high] [--image-only] [--api-key KEY]

Edit existing image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "editing instructions" --filename "output.png" --input-image "path/to/input.png" [--model MODEL] [--resolution 512|1K|2K|4K] [--aspect-ratio RATIO] [--api-key KEY]

Multiple input images (up to 14):

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "combine these elements" --filename "output.png" --input-image "img1.png" "img2.png" "img3.png" [--model MODEL] [--resolution 2K] [--api-key KEY]

Important: Always run from the user's current working directory so images are saved where the user is working, not in the skill directory.

Default Workflow (draft → iterate → final)

Goal: fast iteration without burning time on 4K until the prompt is correct.

• Draft (1K): quick feedback loop
• uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-draft.png" --resolution 1K
• Iterate: adjust prompt in small diffs; keep filename new per run
• If editing: keep the same --input-image for every iteration until you're happy.
• Final (4K): only when prompt is locked
• uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-final.png" --resolution 4K

Model Selection

Use --model to specify the Gemini model. Default: gemini-3.1-flash-image-preview.

Available models:

Map user requests:

• Default / no preference → gemini-3.1-flash-image-preview
• "pro", "best quality", "professional" → gemini-3-pro-image-preview
• "fast", "cheap", "basic" → gemini-2.5-flash-image

Resolution Options

Gemini 3.1 Flash Image supports four resolutions (uppercase K required, except 512):

• 512 (0.5K) - ~512px resolution (fastest, lowest cost)
• 1K (default) - ~1024px resolution
• 2K - ~2048px resolution
• 4K - ~4096px resolution

Map user requests to API parameters:

• No mention of resolution → 1K
• "thumbnail", "tiny", "preview", "0.5K", "512" → 512
• "low resolution", "1080", "1080p", "1K" → 1K
• "2K", "2048", "normal", "medium resolution" → 2K
• "high resolution", "high-res", "hi-res", "4K", "ultra" → 4K

Aspect Ratio Options

14 aspect ratios supported. Use --aspect-ratio to set:

• 1:1 - Square (default if no input image)
• 1:4, 4:1 - Extreme vertical / horizontal (new in 3.1 Flash)
• 1:8, 8:1 - Ultra-extreme vertical / horizontal (new in 3.1 Flash)
• 2:3, 3:2 - Classic portrait / landscape
• 3:4, 4:3 - Standard photo portrait / landscape
• 4:5, 5:4 - Instagram-style portrait / landscape
• 9:16, 16:9 - Phone vertical / widescreen
• 21:9 - Ultra-widescreen / cinematic

Map user requests:

• "square" → 1:1
• "portrait", "vertical" → 3:4 or 9:16
• "landscape", "horizontal" → 4:3 or 16:9
• "widescreen", "cinematic" → 16:9 or 21:9
• "phone", "story", "reel" → 9:16
• "banner", "ultra-wide" → 21:9
• "tall banner", "vertical banner" → 1:4 or 1:8
• "horizontal banner" → 4:1 or 8:1

If no --aspect-ratio is specified, the model defaults to matching the input image's ratio, or 1:1 for text-to-image.

Thinking Level

Control the model's reasoning depth with --thinking-level:

• minimal (default) - Fastest response, lowest latency
• high - Best quality, model reasons more deeply about composition

Use high for complex scenes, detailed compositions, or when quality matters more than speed. The model always uses some thinking internally; minimal just reduces it.

Multiple Input Images

Nano Banana 2 supports up to 14 reference images in a single request:

• Up to 10 images of objects with high-fidelity
• Up to 4 images of characters for character consistency

Use cases:

• Character consistency: Provide a character reference image, generate different poses/angles
• Composition: Combine elements from multiple images into one scene
• Style transfer: Provide a style reference + content image
• Product mockups: Place product on different backgrounds
• 360° views: Generate different angles of a character by providing previous outputs

Example:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py \
  --prompt "An office group photo of these people, they are making funny faces" \
  --input-image person1.png person2.png person3.png person4.png \
  --filename "2026-03-24-10-00-00-group-photo.png" \
  --resolution 2K --aspect-ratio 5:4

Image-Only Mode

Use --image-only to suppress text in the response and return only the generated image. Useful when you don't need the model's text commentary.

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "a sunset" --filename "sunset.png" --image-only

API Key

The script checks for API key in this order:

1. --api-key argument (use if user provided key in chat)
2. GEMINI_API_KEY environment variable

If neither is available, the script exits with an error message.

Preflight + Common Failures (fast fixes)

• Preflight:
• command -v uv (must exist)
• test -n "$GEMINI_API_KEY" (or pass --api-key)
• If editing: test -f "path/to/input.png"

• Common failures:
• Error: No API key provided. → set GEMINI_API_KEY or pass --api-key
• Error loading input image: → wrong path / unreadable file; verify --input-image points to a real image
• Maximum 14 input images supported. → reduce the number of input images
• "quota/permission/403" style API errors → wrong key, no access, or quota exceeded; try a different key/account

Filename Generation

Generate filenames with the pattern: yyyy-mm-dd-hh-mm-ss-name.png

Format: {timestamp}-{descriptive-name}.png

• Timestamp: Current date/time in format yyyy-mm-dd-hh-mm-ss (24-hour format)
• Name: Descriptive lowercase text with hyphens
• Keep the descriptive part concise (1-5 words typically)
• Use context from user's prompt or conversation
• If unclear, use random identifier (e.g., x9k2, a7b3)

Examples:

• Prompt "A serene Japanese garden" → 2025-11-23-14-23-05-japanese-garden.png
• Prompt "sunset over mountains" → 2025-11-23-15-30-12-sunset-mountains.png
• Prompt "create an image of a robot" → 2025-11-23-16-45-33-robot.png
• Unclear context → 2025-11-23-17-12-48-x9k2.png

Image Editing

When the user wants to modify an existing image:

1. Check if they provide an image path or reference an image in the current directory
2. Use --input-image parameter with the path to the image
3. The prompt should contain editing instructions (e.g., "make the sky more dramatic", "remove the person", "change to cartoon style")
4. Common editing tasks: add/remove elements, change style, adjust colors, blur background, inpainting, style transfer, sketch-to-photo, etc.

Prompt Handling

For generation: Pass user's image description as-is to --prompt. Only rework if clearly insufficient.

For editing: Pass editing instructions in --prompt (e.g., "add a rainbow in the sky", "make it look like a watercolor painting")

Preserve user's creative intent in both cases.

Prompt Templates (high hit-rate)

Use templates when the user is vague or when edits must be precise.

• Generation template:
• "Create an image of: . Style: