Name: SAM 2 API: Automatic Image Segmentation by Meta
Brand: magicapi
SKU: api-magicapi-image-segmentation-api-meta-sam-2
Availability: InStock
Author: API Market

Shashank Agarwal

# ✂️ Image Segmentation API (Meta SAM 2)

## Overview
SAM 2 (Segment Anything 2) is a segmentation model from Meta. This listing runs it on images in automatic mode: it samples a grid of points over the whole image and returns a mask for every object or region it finds. There are no point or box prompts.

Send an `image` URL. The response contains `combined_mask`, one image with all masks, and `individual_masks`, a list with one mask image per detected object. Tune how many masks are kept with `points_per_side`, `pred_iou_thresh` and `stability_score_thresh`.

## Features
- **Automatic segmentation**: no clicks, points or boxes needed; the model segments the whole image.
- **Per-object masks**: `individual_masks` returns one mask image per object (31 for the example image).
- **Combined view**: `combined_mask` shows all masks in one image.
- **Sampling density**: `points_per_side` (default 32) sets the grid of points used to find objects.
- **Quality filters**: `pred_iou_thresh` (default 0.88) and `stability_score_thresh` (default 0.95) drop low-quality masks.
- **Mask refinement**: `use_m2m` (default `true`) refines each mask in one extra step.

## Use Cases
1. **Object cut-outs**: use an individual mask to cut an object out of a photo.
2. **Dataset labeling**: generate candidate masks for annotation tools.
3. **Photo editing**: select regions to recolor, blur or replace.
4. **Scene analysis**: count and separate the objects in an image.

## Visual Examples
### Example Input
Image: ![Example input image sent to SAM 2](https://blog.api.market/wp-content/uploads/2026/10/sam-2-input-api-market.jpg)

All settings were at their defaults: `points_per_side` `32`, `pred_iou_thresh` `0.88`, `stability_score_thresh` `0.95`, `use_m2m` `true`.

### Example Output
Combined mask:

![SAM 2 combined mask for the example image](https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-1-9pvx7z1w-api-market.jpg)

The request returned 31 individual masks. The first three:

![SAM 2 individual mask 1](https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-2-9pvx7z1w-api-market.png)
![SAM 2 individual mask 2](https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-3-9pvx7z1w-api-market.png)
![SAM 2 individual mask 3](https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-4-9pvx7z1w-api-market.png)

All 31 mask URLs are listed in the 200 response below.

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `image` | string (URL) | Yes |  |  | Image URL to segment |
| `points_per_side` | integer | No | `32` |  | Points per side of the sampling grid used for automatic mask generation |
| `pred_iou_thresh` | number | No | `0.88` |  | Predicted IoU threshold: masks with a lower predicted quality score are discarded |
| `stability_score_thresh` | number | No | `0.95` |  | Stability score threshold: masks with a lower stability score are discarded |
| `use_m2m` | boolean | No | `true` |  | Use M2M, a one-step refinement of each mask using the previous mask prediction |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.03 per API unit.
- **1 API unit = 1 request covering up to 27 seconds of processing.** A request that processes for longer uses one unit per started 27-second block (for example 30 seconds = 2 units).
- The number of masks returned does not change the price.
- **Example cost**: the example request processed for 10.14 seconds = 1 API unit = **$0.03**.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/image-segmentation-api-meta-sam-2`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 1 second and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/image-segmentation-api-meta-sam-2/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "fe97b453a6455861e3bac769b441ca1f1086110da7466dbb65cf1eecfd60dc83",
    "input": {
      "image": "https://blog.api.market/wp-content/uploads/2026/10/sam-2-input-api-market.jpg",
      "use_m2m": true,
      "points_per_side": 32,
      "pred_iou_thresh": 0.88,
      "stability_score_thresh": 0.95
    }
  }'
```

Response (201 Created):
```json
{
  "id": "9pvx7z1wcxrmw0d1c6naymx09g",
  "version": "fe97b453a6455861e3bac769b441ca1f1086110da7466dbb65cf1eecfd60dc83",
  "status": "starting",
  "created_at": "2026-10-22T03:31:49.172Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/image-segmentation-api-meta-sam-2/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "9pvx7z1wcxrmw0d1c6naymx09g",
  "status": "succeeded",
  "output": {
    "combined_mask": "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-1-9pvx7z1w-api-market.jpg",
    "individual_masks": [
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-2-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-3-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-4-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-5-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-6-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-7-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-8-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-9-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-10-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-11-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-12-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-13-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-14-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-15-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-16-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-17-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-18-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-19-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-20-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-21-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-22-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-23-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-24-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-25-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-26-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-27-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-28-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-29-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-30-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-31-9pvx7z1w-api-market.png",
      "https://blog.api.market/wp-content/uploads/2026/10/sam-2-output-32-9pvx7z1w-api-market.png"
    ]
  },
  "error": null,
  "created_at": "2026-10-22T03:31:49.172Z",
  "started_at": "2026-10-22T03:31:50.172Z",
  "completed_at": "2026-10-22T03:32:00.312Z"
}
```

`output` is an object: `combined_mask` is the URL of one image showing all masks, and `individual_masks` is a list of URLs, one mask image per detected object.

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a missing `image` or a `points_per_side` value that is not an integer.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits
All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.
## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/image-segmentation-api-meta-sam-2";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "fe97b453a6455861e3bac769b441ca1f1086110da7466dbb65cf1eecfd60dc83", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 1000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output;
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

createAndWaitForPrediction({
  "image": "https://blog.api.market/wp-content/uploads/2026/10/sam-2-input-api-market.jpg",
  "use_m2m": true,
  "points_per_side": 32,
  "pred_iou_thresh": 0.88,
  "stability_score_thresh": 0.95
}).then((output) => console.log(output));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/image-segmentation-api-meta-sam-2"


def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "fe97b453a6455861e3bac769b441ca1f1086110da7466dbb65cf1eecfd60dc83", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])


def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 1
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return prediction["output"]

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")


output = create_and_wait_for_prediction({
    "image": "https://blog.api.market/wp-content/uploads/2026/10/sam-2-input-api-market.jpg",
    "use_m2m": True,
    "points_per_side": 32,
    "pred_iou_thresh": 0.88,
    "stability_score_thresh": 0.95
})
print(output)
```

## FAQ
- **Can I click a point or draw a box to segment one object?**
  - No. This listing has no point or box inputs; it segments the whole image automatically. Pick the mask you need from `individual_masks`.
- **How do I get fewer, cleaner masks?**
  - Raise `pred_iou_thresh` or `stability_score_thresh` to discard lower-quality masks, or lower `points_per_side` to sample fewer points.
- **How do I get more masks for small objects?**
  - Raise `points_per_side` to sample a denser grid, or lower the two thresholds. Denser grids take longer to process.
- **How much does a request cost?**
  - $0.03 for each request that processes in 27 seconds or less, regardless of how many masks it returns. The example took 10.14 seconds.
- **Is the 201 response from POST an error?**
  - No. 201 Created means the prediction was accepted. Use the returned `id` to poll GET `/predictions/{id}` until `status` is `succeeded`, `failed` or `canceled`.

## Contact Us
If you have any questions or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


SAM 2 API: segment every object in an image automatically with Meta's Segment Anything 2 and get a combined mask plus one mask per object. Pay per image.