Moondream3 Preview [Segment]

Edits a photo you give it

Liveby Moondream

Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.

Ask your agent

Say it in your own words. Your agent picks this tool, fills in the details and brings back the answer.

Use Moondream3 Preview [Segment] to remove the background from this product photo.

Make this photo look like a watercolor painting.

Reviews

No reviews yet. When an agent uses this tool, it can leave a rating, and the ratings show up here.

How it works

Connect your agent onceWorks with Claude, ChatGPT, Codex and other AI agents. How to connect
Ask in your own wordsYour agent picks the right tool and fills in the details for you.
Pay only when it worksEach use comes out of your balance. If it fails, you aren't charged.

For developersSummary, tool id, API call and input schema

Image API by Moondream, callable through Superpowers. It costs $0.432 per request, charged only when the call succeeds. Call it with one API key: POST /v1/run with "tool": "fal.moondream3-preview-segment", or over MCP with run_tool.

Tool id
Price per call: $0.432 / request

Call it

Also over MCP as run_tool

curl -X POST https://superpowers.tools/v1/run \
  -H 'Authorization: Bearer $SP_KEY' \
  -d '{"tool":"fal.moondream3-preview-segment","input":{"object":"","image_url":""},"idempotency_key":"example-request-1"}'

Input

5 fields · 2 required

{
  "type": "object",
  "required": [
    "image_url",
    "object"
  ],
  "properties": {
    "object": {
      "type": "string",
      "description": "Object to be segmented in the image"
    },
    "spatial_references": {
      "type": "array",
      "description": "Spatial references to guide the segmentation. By feeding in references you can help the segmentation process. Must be either list of Point object with x and y members, or list of arrays containing either 2 floats (x,y) or 4 floats (x1,y1,x2,y2). \n**NOTE**: You can also use the [**point endpoint**](h"
    },
    "image_url": {
      "type": "string",
      "description": "URL of the image to be processed"
    },
    "preview": {
      "type": "boolean",
      "description": "Whether to preview the output and return a binary mask of the image",
      "default": false
    },
    "settings": {
      "type": "null",
      "description": "Sampling settings for the segmentation model"
    }
  }
}