Nano Banana Pro Image Generation & Editing
Quick Start
Generate:
uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "A serene Japanese garden" --filename "2025-11-23-14-23-05-japanese-garden.png" --resolution 1K
Edit:
uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "make the sky dramatic" --filename "2025-11-23-14-25-30-dramatic-sky.png" --input-image "original.png" --resolution 2K
Compose:
uv run ~/.codex/skills/nano-banana-pro/scripts/compose_images.py "Create a group photo" group.png person1.png person2.png
Multi-turn chat:
uv run ~/.codex/skills/nano-banana-pro/scripts/multi_turn_chat.py --model gemini-3-pro-image-preview --output-dir .
Scripts
generate_image.py: text-to-image + edit via --input-image, resolution 1K/2K/4K, optional --aspect, auto-detect resolution on edit.edit_image.py: explicit edit CLI (input, instruction, output).compose_images.py: combine up to 14 reference images.multi_turn_chat.py: interactive refine session (/save, /load, /clear, /quit).gemini_images.py: Python helper class.
Model
- Default:
gemini-3-pro-image-preview.
- Only use other models if user asks.
Resolution + Aspect
- Resolution:
1K default, 2K, 4K.
- Aspect ratios (where supported):
1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.
- Map words: "1080/low/1K"->1K, "2K/2048/medium"->2K, "4K/high/ultra"->4K.
Workflow
- Draft 1K, iterate small prompt diffs, final 4K.
- For edits, keep same
--input-image until final.
API Key
GEMINI_API_KEY required.generate_image.py also supports --api-key.
Filenames
- Pattern:
yyyy-mm-dd-hh-mm-ss-name.png.
- Lowercase, hyphenated, 1-5 words.
Output
- Run from user cwd so files save there.
generate_image.py converts output to PNG.- Do not open/read images; report saved path.
Preflight
command -v uvtest -n "$GEMINI_API_KEY"- If editing:
test -f "path/to/input.png"
Common Failures
- No API key.
- Input image missing/unreadable.
- Quota/permission/403 errors.
1---2name: nano-banana-pro3description: Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image creation or modification incl logos, stickers, mockups, style transfer, multi-image composition, and multi-turn refinement.4---56# Nano Banana Pro Image Generation & Editing78## Quick Start9Generate:10```bash11uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "A serene Japanese garden" --filename "2025-11-23-14-23-05-japanese-garden.png" --resolution 1K12```1314Edit:15```bash16uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "make the sky dramatic" --filename "2025-11-23-14-25-30-dramatic-sky.png" --input-image "original.png" --resolution 2K17```1819Compose:20```bash21uv run ~/.codex/skills/nano-banana-pro/scripts/compose_images.py "Create a group photo" group.png person1.png person2.png22```2324Multi-turn chat:25```bash26uv run ~/.codex/skills/nano-banana-pro/scripts/multi_turn_chat.py --model gemini-3-pro-image-preview --output-dir .27```2829## Scripts30- `generate_image.py`: text-to-image + edit via `--input-image`, resolution 1K/2K/4K, optional `--aspect`, auto-detect resolution on edit.31- `edit_image.py`: explicit edit CLI (input, instruction, output).32- `compose_images.py`: combine up to 14 reference images.33- `multi_turn_chat.py`: interactive refine session (`/save`, `/load`, `/clear`, `/quit`).34- `gemini_images.py`: Python helper class.3536## Model37- Default: `gemini-3-pro-image-preview`.38- Only use other models if user asks.3940## Resolution + Aspect41- Resolution: `1K` default, `2K`, `4K`.42- Aspect ratios (where supported): `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`.43- Map words: "1080/low/1K"->1K, "2K/2048/medium"->2K, "4K/high/ultra"->4K.4445## Workflow46- Draft 1K, iterate small prompt diffs, final 4K.47- For edits, keep same `--input-image` until final.4849## API Key50- `GEMINI_API_KEY` required.51- `generate_image.py` also supports `--api-key`.5253## Filenames54- Pattern: `yyyy-mm-dd-hh-mm-ss-name.png`.55- Lowercase, hyphenated, 1-5 words.5657## Output58- Run from user cwd so files save there.59- `generate_image.py` converts output to PNG.60- Do not open/read images; report saved path.6162## Preflight63- `command -v uv`64- `test -n "$GEMINI_API_KEY"`65- If editing: `test -f "path/to/input.png"`6667## Common Failures68- No API key.69- Input image missing/unreadable.70- Quota/permission/403 errors.
marchatton