Generate transparent PNG images from text prompts using Nano Banana Pro + background removal
Project description
nanopng
CLI to generate transparent PNG images from text prompts. Describe what you want, get a cutout PNG in ~12 seconds.
Install
uv tool install nanopng
On first run, nanopng will prompt for your Gemini API key and save it to ~/.nanopng/.env. Get a key at aistudio.google.com/apikey.
Usage
nanopng "fire breathing dragon"
nanopng "cat wizard" -n 4
nanopng "samurai warrior" --style anime --size 9:16
nanopng "crystal sword" --model standard -o sword.png
# With reference image (uses Gemini 3.1 Flash Image)
nanopng "character in this style" -i reference.png -o output.png
# Multiple reference images (up to 14)
nanopng "combine these elements" -i img1.png -i img2.png -i img3.png
# From stdin (for piping image data)
cat reference.jpg | nanopng -i - "prompt" -o output.png
Claude Code Integration
When using nanopng with Claude Code, you can send images directly in chat:
- Send an image in the conversation
- Claude finds it in
/var/folders/.../T/tos_*.png - Pipes it directly:
cat [temp-file] | nanopng -i - "prompt" - Instant generation with your reference image!
Update
uv tool upgrade nanopng
Options
| Flag | Description | Default |
|---|---|---|
-o, --output |
Output file path | Auto from prompt |
-n |
Number of images (1-4) | 1 |
--size |
Aspect ratio: 1:1, 16:9, 9:16, 4:3, 3:4 |
1:1 |
--style |
Style preset (see below) | None |
--model |
fast (~10s) or standard (~17s) |
fast |
Styles
pixel-art watercolor 3d-render photorealistic cartoon anime oil-painting flat
Tech Stack
| Component | Tools |
|---|---|
| Language | Python 3.12 |
| Image Gen | Google Imagen 4 Fast |
| BG Removal | rembg (isnet-general-use) |
| CLI | argparse (stdlib) |
Architecture
[User Input] [System Boundary: nanopng] [External]
┌──────────────┐ ┌───────────────┐
│ Terminal │ │ Google API │
│ │ │ (Imagen 4) │
│ $ nanopng │ └───────┬───────┘
│ "prompt" │ │
└──────┬───────┘ │ HTTPS
│ │
│ CLI Command │
▼ │
┌─────────────────────────────────────────────────────────┐ │
│ Python CLI (argparse) │ │
│ │ │
│ ┌──────────────────────────────────────────────────┐ │ │
│ │ cli.py │ │ │
│ │ │ │ │
│ │ Parse args (prompt, -n, --size, --style, etc.) │ │ │
│ └──────────────┬───────────────────────────────────┘ │ │
│ │ │ │
│ ▼ │ │
│ ┌──────────────────────────────────────────────────┐ │ │
│ │ core.py │ │◀────────┘
│ │ │ │ Generate
│ │ 1. Build prompt + chroma key suffix │ │
│ │ 2. Call Imagen 4 Fast (google-genai) │ │
│ │ 3. Remove background (rembg, local) │ │
│ │ 4. Save RGBA transparent PNG │ │
│ └──────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────┘
│
▼
prompt-slug.png
(transparent RGBA)
Imagen 4 Fast (~10s) • rembg local inference (~2s) • ~12s total
Storage
~/.nanopng/.env # API key (saved on first run)
~/.u2net/isnet-general-use.onnx # BG removal model (one-time download, ~179MB)
License
MIT
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file nanopng-0.2.0.tar.gz.
File metadata
- Download URL: nanopng-0.2.0.tar.gz
- Upload date:
- Size: 6.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.8.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
282f3ef1304762015e9940204377575c8a28b56cc72cb2f93b9cd1bd0e84a928
|
|
| MD5 |
a802810d457b0543de6e5487c144eb5b
|
|
| BLAKE2b-256 |
733c2c4fabe7463697976dc7a5960f15185bd661075f6ee7bd6ab3ab02b26042
|
File details
Details for the file nanopng-0.2.0-py3-none-any.whl.
File metadata
- Download URL: nanopng-0.2.0-py3-none-any.whl
- Upload date:
- Size: 6.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.8.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
befdfaf9c1b039caa22b6742e51117414d17a2e563a44408edc9171a6db1be91
|
|
| MD5 |
a49dd4bdf4ee6ba4da468dbf9e29cfcf
|
|
| BLAKE2b-256 |
9bdd2dd2f24b0b83114e4dafec17c667af8431f26585887dcd4a6c9125c32b71
|