How to Use SAM 3.1 Image in Claude
SAM 3.1 Image extracts object masks from photos using short descriptions, points, or boxes. Use it for product cutouts, editing selections, and draft annotations.
Prompts to try with SAM 3.1 Image in Claude
Copy an example into Claude, or ask in your own words.
Use SAM 3.1 Image to select the "red backpack" in this photo. Show the mask over the original so I can check the outline.
Use SAM 3.1 Image to select "chair" in this room photo. Return separate selections and help me check whether any chairs were missed.
Use SAM 3.1 Image to select the jacket in this portrait. Prepare its mask for a separate color edit.
Use SAM 3.1 Image to isolate the lamp inside the box I marked on this photo. Show the selection before preparing a cutout.
Use SAM 3.1 Image to select "screen" in this photo. Show every returned selection so I can review the areas for a later blur step.
How to set up SAM 3.1 Image 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 SAM 3.1 Image
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": "fal-ai/sam-3-1/image", "input": { "image_url": "https://raw.githubusercontent.com/facebookresearch/segment-anything-2/main/notebooks/images/truck.jpg" } }'
Then paste one of the prompts above: Claude finds SAM 3.1 Image on Spiral, shows the price and runs it.
SAM 3.1 Image providers and prices
Spiral runs SAM 3.1 Image through fal. Open a card for today's price, with every fee included.
Settings you can ask for
SAM 3.1 Image works from your image URL and prompt. Ask Claude for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Apply mask | on or off | on | Apply the mask on the image. |
| Output format | jpeg, png, or webp | png | The format of the generated image. |
| Return multiple masks | on or off | off | If True, upload and return multiple generated masks as defined by |
| Max masks | 1 to 32 | 3 | Maximum number of masks to return when |
| Include scores | on or off | off | Whether to include mask confidence scores. |
| Include boxes | on or off | off | Whether to include bounding boxes for each mask (when available). |
Every API field of Sam 3 1 · Image to Image is on its tool page: open its card above.
What can you do with SAM 3.1 Image in Claude?
Turn a shared photo into selections for cutouts, edits, or object review. A chat app can help choose a short description and show the resulting masks against the original image.
Product cutout preparation
Select a shoe, bag, or other product in a photo, then use its mask to prepare a cutout. Inspect the selection before placing it on a new background.
Selective photo edits
Extract a mask for a jacket, chair, or car so an image editor can apply a color or exposure change to that object.
Draft segmentation annotations
Generate object masks for a labeled image dataset. Have an annotator correct the selections before accepting them as training labels.
Object crop preparation
Locate objects in a photo and use their bounding boxes to prepare individual crops for a catalog or review sheet.
Count review
Request separate selections for a visible category, such as chairs or bottles. Compare the masks with the photo to produce a reviewed count.
Redaction preparation
Select visible screens or license plates to prepare masks for a separate blur or pixelation step. Review the full photo for missed regions before sharing.
More on SAM 3.1 Image and Claude
Overview
SAM 3.1 Image in other agents
Questions about SAM 3.1 Image in Claude
Can Claude use SAM 3.1 Image?
Yes. Add Spiral to Claude as an MCP server, and Claude can find and run SAM 3.1 Image like any built-in tool.
How much does SAM 3.1 Image cost in Claude?
$0.013 per run at the default inputs. 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 SAM 3.1 Image in Claude?
No. Spiral runs SAM 3.1 Image on its own fal account. You sign in to Spiral once and never paste a key.
How long does it take to set up SAM 3.1 Image in Claude?
About two minutes: add Spiral to Claude, then keep the detected sign-in and click Add, then click Connect.
Does Claude wait while SAM 3.1 Image runs?
SAM 3.1 Image runs in the background. Claude starts the run, checks on it and gives you the image when it's ready.
What is SAM 3.1 Image useful for?
Use it to select objects in existing photos. The resulting masks can support product cutouts, selective edits, and draft image annotations.
How should I describe the object I want?
Start with a short phrase such as "red backpack" or "car wheel." Meta documents limitations with longer instructions involving complex spatial relationships, so use a point or box to clarify the target.
Can SAM 3.1 select several objects of the same type?
Concept segmentation targets all matching instances, such as the chairs in a room. Review the returned masks before using them as a count or completed annotation.
What size are the segmentation masks?
Meta's reference implementation resizes binary masks to the original image dimensions. Use those dimensions when aligning a mask with the source photo.
Can I correct a selection with points?
Yes. Points associated with the same object refine its selection; use them when the text description leaves the target ambiguous.
Start using SAM 3.1 Image 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


