fal-image-plugin

KO | EN

gitlite — search

#javascript #python #hacktoberfest #android #machine-learning #deep-learning #macos #awesome #cli #kubernetes #vue #awesome-list #pytorch #openai

fal-image-plugin

(★ 1)

Claude Code plugin for AI image generation via fal.ai and Nano Banana

fal-image-plugin 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# fal-image — Claude Code Plugin

AI image generation for [Claude Code](https://claude.ai/code) via [fal.ai](https://fal.ai), optimized for **Nano Banana 2** (Gemini 3.1 Flash Image).

## Installation

```bash
# Add the marketplace
/plugin marketplace add dubtor/dubtor-plugins

# Install the plugin
/plugin install fal-image@dubtor-plugins
```

## Setup

You need a fal.ai API key:

1. Create an account at [fal.ai](https://fal.ai)
2. Get your key at [fal.ai/dashboard/keys](https://fal.ai/dashboard/keys)
3. Set the environment variable:

```bash
export FAL_KEY=your_key_here
```

## Usage

```bash
# Text-to-image — Claude enhances your prompt automatically
/fal-image:fal-image "A blazer on linen fabric"

# With options
/fal-image:fal-image "A modern office" --ratio 16:9 --resolution 2K --count 4

# Send prompt as-is (no enhancement)
/fal-image:fal-image "Your detailed prompt here..." --no-enhance

# Image-to-image editing with reference images
/fal-image:fal-image "Make this brighter" --reference photo.png
/fal-image:fal-image "Combine these" --reference style.png --reference content.png
```

## Options

| Flag | Description | Default |
|---|---|---|
| `--count N`, `-n N` | Number of images (1-4) | 2 |
| `--ratio RATIO`, `-r RATIO` | Aspect ratio (1:1, 16:9, 9:16, 3:2, etc.) | 1:1 |
| `--resolution RES` | 0.5K, 1K, 2K, or 4K | 1K |
| `--format FMT`, `-f FMT` | png, jpeg, or webp | png |
| `--output DIR`, `-o DIR` | Output directory | ./generated-images |
| `--name NAME` | Filename prefix | Auto from prompt |
| `--reference FILE` | Reference image (repeatable) | — |
| `--thinking LEVEL` | Model reasoning: minimal or high | — |
| `--seed N` | Random seed for reproducibility | — |
| `--no-enhance` | Skip prompt optimization | — |
| `--quiet`, `-q` | Minimal output | — |

## Model

This plugin is optimized for **Nano Banana 2** (`fal-ai/nano-banana-2`), powered by Gemini 3.1 Flash Image. It supports:

- **Text-to-Image**: `fal-ai/nano-banana-2` — generate images from text prompts
- **Image-to-Image**: `fal-ai/nano-banana-2/edit` — edit or transform existing images using reference photos (activated by `--reference`)

Key capabilities of Nano Banana 2:
- Deep reasoning for accurate prompt interpretation
- Real-time web search for up-to-date visual references
- Text rendering in 10+ languages
- Native support for resolutions from 0.5K to 4K
- Up to 14 reference images per request

## Prompt Enhancement

By default, Claude enhances your prompt with composition, lighting, camera/lens, and color grading details — based on the [Google Nano Banana Prompting Guide](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-nano-banana). Use `--no-enhance` to skip this.

## Pricing

~$0.08 per image at 1K resolution (as of March 2026). See [fal.ai/pricing](https://fal.ai/pricing) for current rates.

## Requirements

- Python 3 (standard library only, no pip packages needed)
- `FAL_KEY` environment variable

## Author

Robert Clemens — [83 Ventures GmbH](https://83ventures.io) · [@dubtor](https://github.com/dubtor)

## License

MIT

fal-image-plugin

fal-image-plugin 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

fal-image-plugin

fal-image-plugin 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README