# Image Generation through toolfare

> Images from a text prompt, optionally guided by reference images, or edits to images you already have — from gpt-image-2, gpt-image-2.5, ideogram-v4 or nano-banana-2, asked from Claude, ChatGPT, Codex, Cursor or your own script. No model account or key of your own: toolfare quotes each call before it runs and charges one balance, and the image comes back in the same call.

Page: https://toolfare.ai/image-generation

## Use it from your agent

Paste this into Claude Code, Codex, Cursor or any agent with a shell:

> Set up https://toolfare.ai/skill.md, then use toolfare to generate a 4:5 image of a matte black water bottle on a sunlit concrete step with nano-banana-2, and tell me the price before you start

Apps without a terminal connect toolfare as an MCP server: https://toolfare.ai/docs#connect

## Use it from a script

```sh
mkdir -p ~/.local/bin
curl -fsSL 'https://toolfare.ai/toolfare/toolfare.mjs' -o ~/.local/bin/toolfare
chmod +x ~/.local/bin/toolfare
export PATH="$HOME/.local/bin:$PATH"
export TOOLFARE_MCP_URL='https://api.toolfare.ai/mcp'
toolfare login
```

Sign in once per machine; the CLI renews the sign-in itself. On a CI runner or another server,
create an API key in the toolfare console and set it as `TOOLFARE_API_KEY` instead of signing in.
Paid calls on one sign-in or key run one at a time. `--max-cost` refuses a call quoted above
that many credits (1,000 credits = $3).

## Tools

### An image from a prompt, with optional references

Tool: `image_generation`

Text-to-image generation, optionally from input images

Required parameters: model, prompt

Optional parameters: aspect_ratio, image, n, output_format, quality, resolution, response_format, size

Cost: From $0.09

Ask:
- Make a square hero image of a ceramic coffee cup on a linen tablecloth.
- Generate a 9:16 story background for our summer sale, with space for text at the top.
- Draw a poster with the words “Open late on Fridays” using ideogram-v4.

### Change an image you already have

Tool: `image_generation_edit`

Image editing from input images

Required parameters: image, model, prompt

Optional parameters: aspect_ratio, n, output_format, quality, resolution, response_format, size

Cost: From $0.24

Ask:
- Put this product photo on a marble countertop instead of the white background.
- Change the jacket in this image from red to navy.
- Combine these three product shots into one flat-lay image.
## Price

From $0.09 Quoted before each call; the final charge follows the returned usage. Pay per call from one balance.

Balances and top-ups are in US dollars. Quotes and `--max-cost` count credits: 1,000 credits = $3 ($0.003 each).

## Use it when

- You need a still image for an ad, a post, a landing page or a mock-up.
- You want to compare how different models handle the same prompt before choosing one.
- You have a product photo and want it in a new scene or colourway.
- Your agent should produce images inside a script or pipeline, not a design tool.

## Look elsewhere when

- You specifically want ByteDance’s Seedream models, or output sizes up to 4K set by width and height.
- You only need a mechanical fix — remove a background, upscale, outpaint or erase an object.
- You want moving pictures rather than a still.

## Limits

- One image per call; ask again for variations.
- Options depend on the model: aspect ratios apply to ideogram-v4 and nano-banana-2, resolution up to 4K to nano-banana-2, and quality and size to the gpt-image models.
- Reference and edit inputs are image URLs or data URLs — up to 4 for the gpt-image models and up to 14 for nano-banana-2. ideogram-v4 takes no input images.
- Pay as you go from your toolfare balance, in US dollars; there is no subscription. A call runs when your balance covers its quote.