How to Use GPT Image 2 Edit in Claude
Use GPT Image 2 Edit to revise an existing image, combine references, or change a scene while preserving its subject. It uses the same underlying model as GPT Image 2, with reference images guiding the result.
Prompts to try with GPT Image 2 Edit in Claude
Copy an example into Claude, or ask in your own words.
Use GPT Image 2 Edit on my reference photograph. Change only the wall color to muted green. Preserve the furniture, camera angle, window light, and floor shadows. Return a 1536×1024 image.
Use GPT Image 2 Edit to place the bicycle from the second reference into the courtyard in the first reference. Match its scale, perspective, and shadows to the courtyard. Preserve the courtyard framing and return a 1536×1024 image.
Remove the background from the photo I attach with GPT Image 2 Edit and give me a transparent PNG.
Use GPT Image 2 Edit to clean up these product shots so they sit on pure white.
Edit this image with GPT Image 2 Edit: keep the subject, swap the background for a sunlit studio.
How to set up GPT Image 2 Edit in Claude
Pick the way that suits you. Each takes about two minutes.
Use the Claude connector
- Add Spiral to Claude
Click Add Spiral in Claude above. Claude opens Add custom connector with Spiral and its URL filled in; check them and click Continue. Without the link: in Claude open Customize › Connectors › Add › Add custom connector and paste this URL:
https://api.spiral.ai/mcp
- Keep the detected sign-in and click Add
Claude finds Spiral's sign-in by itself: leave Sign in now and Use Claude's published identity selected.
- Click Connect
The connector page opens. Connect takes you to Spiral in your browser.
Use the Spiral skill
- Send Claude this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run GPT Image 2 Edit
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "fal", "model": "openai/gpt-image-2/edit", "input": { "prompt": "Same workers, same beam, same lunch boxes - but they'\''re all on their phones now. One is taking a selfie. One is on a call looking annoyed. Same danger, new…", "image_urls": [ "<image_urls>" ] } }'
Then paste one of the prompts above: Claude finds GPT Image 2 Edit on Spiral, shows the price and runs it.
GPT Image 2 Edit providers and prices
Spiral runs GPT Image 2 Edit through fal and OpenAI. Open a card for today's price, with every fee included. Name one in your prompt, or let Claude pick.
$0.188 per 4:3 image at high quality ($0.00572 low, $0.0475 medium). At high: $0.24 square, $0.274 square HD, $0.135 16:9. Auto quality or size: the price of the one chosen for the request. Prompt: $6.50 per 1M tokens. Input images: $0.0106 each at 1024×1024.
$0.274 per 1024×1024 image at high quality ($0.00775 low, $0.0686 medium), plus each input image. At high: $0.214 at 1536×1024 or 1024×1536, $0.198 at 1792×1024 or 1024×1792, $0.203 at 512×512, $0.186 at 256×256. Auto quality or size: the price of the one chosen for the request. Prompt: $6.50 per 1M tokens. Input images: $0.0106 each at 1024×1024.
Settings you can ask for
GPT Image 2 Edit works from your prompt, image URLs, and mask URL. Ask Claude for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Background | auto or opaque | auto | Background for the generated image |
| Quality | auto, low, medium, or high | high | Quality for the generated image. Use 'auto' to let the model pick the best quality for the prompt. |
| Number of images | 1 to 10 | 1 | Number of images to generate |
| Output format | jpeg, png, or webp | png | Output format for the images |
| Output compression | 0 to 100 | Compression level from 0 to 100. Only supported when output_format is 'jpeg' or 'webp'. | |
| Image passthrough | on or off | off | Send reference images and masks without resizing or re-encoding, and return the provider's original image bytes and format. Requests use the direct OpenAI API. Masks must be PNGs with an alpha channel: transparent regions are edited, and dimensions must match the first reference. |
Every API field of GPT Image 2 · Image Editing is on its tool page: open its card above.
What can you do with GPT Image 2 Edit in Claude?
Turn a written idea into a poster, diagram, or realistic scene you can review in conversation. With reference images, you can request a focused revision and refine one detail at a time.
Article illustrations
Turn an article brief into a realistic scene or explanatory illustration, with space reserved for a headline.
Labeled diagrams
Create an infographic from supplied relationships and exact labels, then review the text and arrangement.
Product scene revisions
Edit a product photograph into a different setting while specifying which shape, label, and lighting details to preserve.
Visual sequences
Describe each panel of a short visual story and review the result for subject continuity.
Campaign concepts
Explore poster compositions at low quality, then refine a chosen direction with exact copy at higher quality.
Reference composites
Combine a subject from one image with a scene from another, naming each reference's role.
GPT Image 2 Edit success rate and speed on Spiral
Success and speed over the last 24 hours, across every run of these tools on Spiral.
| Tool | Success | Median time | Runs |
|---|---|---|---|
| GPT Image 2 · Image Editing | 96.0% | 1m 32s | 274 |
| GPT Image 2 · Edit | Not enough runs | ||
More on GPT Image 2 Edit and Claude
Overview
GPT Image 2 Edit in other agents
Questions about GPT Image 2 Edit in Claude
Can Claude use GPT Image 2 Edit?
Yes. Add Spiral to Claude as an MCP server, and Claude can find and run GPT Image 2 Edit like any built-in tool.
How much does GPT Image 2 Edit cost in Claude?
$0.188 per 4:3 image at high quality ($0.00572 low, $0.0475 medium). At high: $0.24 square, $0.274 square HD, $0.135 16:9. Auto quality or size: the price of the one chosen for the request. Prompt: $6.50 per 1M tokens. Input images: $0.0106 each at 1024×1024. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude plan is separate.
Do I need a fal account to use GPT Image 2 Edit in Claude?
No. Spiral runs GPT Image 2 Edit on its own fal account. You sign in to Spiral once and never paste a key.
How long does it take to set up GPT Image 2 Edit in Claude?
About two minutes: add Spiral to Claude, then keep the detected sign-in and click Add, then click Connect.
Which provider runs GPT Image 2 Edit?
Spiral offers GPT Image 2 Edit through fal and OpenAI. Name one in your prompt, or let Claude pick.
Does Claude wait while GPT Image 2 Edit runs?
GPT Image 2 Edit runs in the background. Claude starts the run, checks on it and gives you the image when it's ready.
What is GPT Image 2 good at?
It is useful for realistic imagery, posters with embedded text, infographics, diagrams, and visual sequences. Give it a concrete brief with the subject, composition, required copy, and intended use.
How does GPT Image 2 Edit differ from GPT Image 2?
Both use the same underlying model. Choose the generation entry to start from a written brief and the editing entry to revise existing imagery or combine references.
Can GPT Image 2 generate 4K images?
Yes. Supported examples include 3840×2160 landscape and 2160×3840 portrait; custom sizes must meet the dimension, aspect ratio, and total pixel limits.
How do I get readable text inside an image?
Quote the exact copy, specify its position and typography, and say how many times it should appear. Check the result for spelling and placement before using it.
Which quality setting should I use?
Start with low quality to compare draft concepts. Use medium or high for final assets, especially when small lettering or dense labels matter.
Start using GPT Image 2 Edit in Claude
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Claude:
set up https://spiral.ai/SKILL.md


