Name: Glance Qwen3-VL-4B API: Image Q&A with Probabilities
Brand: magicapi
SKU: api-magicapi-typed-answers-api-untapped-glance-qwen3-vl-4b
Availability: InStock
Author: API Market

Shashank Agarwal

# ✅ Image Question Answering API (untapped Glance Qwen3-VL-4B)

## Overview
Glance answers a yes/no or multiple-choice question about an image and returns the answer with probabilities. Its developer describes it as typed answers about an image, read from unchanged Qwen3-VL-4B logits in one forward pass.

Send a `question`, an image (`image` URL or `image_base64`) and, for multiple choice, a list of options in `options_json`. Instead of free text, the model returns a JSON object with the chosen `answer`, a `confidence` value and a probability for every option.

## Features
- **Two question types**: `yes_no` (default) or `choice`.
- **Your own options**: for `choice`, pass the labels as a JSON array string in `options_json`, for example `["Cat","Dog","Bird","Horse"]`.
- **Probabilities per option**: every option comes back with a probability, so you can set your own thresholds.
- **Image by URL or base64**: JPEG, PNG or WebP via `image`, or a base64 string via `image_base64`.
- **Structured output**: one JSON object, no text parsing needed.

## Use Cases
1. **Image classification with your own labels**: product category, animal type, document type.
2. **Content checks**: yes/no questions such as whether an image contains a person or text.
3. **Form and listing validation**: confirm a photo matches what a user selected.
4. **Routing and tagging**: send images to the right queue based on the chosen option and its confidence.

## Visual Examples
### Example Input
![Glance example input image](https://blog.api.market/wp-content/uploads/2026/10/glance-qwen3-vl-4b-input-api-market.jpg)

- `question`: "Which animal is shown in this image?"
- `question_type`: `choice`
- `options_json`: `["Cat","Dog","Bird","Horse"]`

### Example Output
```json
{
  "answer": "Cat",
  "confidence": 1,
  "probabilities": [
    { "label": "Cat", "probability": 1 },
    { "label": "Dog", "probability": 0 },
    { "label": "Bird", "probability": 0 },
    { "label": "Horse", "probability": 0 }
  ]
}
```

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `question` | string | Yes |  |  | Question about the image |
| `image` | string (URL) | No | none |  | JPEG, PNG, or WebP image |
| `image_base64` | string | No | `""` |  | Base64-encoded image for API clients |
| `question_type` | string | No | `yes_no` | `yes_no`, `choice` | Type of question |
| `options_json` | string | No | `[]` | JSON array of strings, sent as a string | JSON array of choice labels |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.018 per API unit.
- **1 API unit = 1 successful request with up to 72 seconds of processing.** A longer job bills one more unit for each started block of 72 seconds.
- **Example cost**: the example request took 46.2 seconds of processing, which is 1 unit, so $0.018.
- **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/typed-answers-api-untapped-glance-qwen3-vl-4b`

### 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/typed-answers-api-untapped-glance-qwen3-vl-4b/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "65c82d4fbfad89eaee2dabd4abd0be8703d7a3187e216edbde26eecc6f56d69f",
    "input": {
      "image": "https://blog.api.market/wp-content/uploads/2026/10/glance-qwen3-vl-4b-input-api-market.jpg",
      "question": "Which animal is shown in this image?",
      "options_json": "[\"Cat\",\"Dog\",\"Bird\",\"Horse\"]",
      "question_type": "choice"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "qqvtd6nwwdrgc0d1d8wth7hp60",
  "version": "65c82d4fbfad89eaee2dabd4abd0be8703d7a3187e216edbde26eecc6f56d69f",
  "status": "starting",
  "created_at": "2026-10-23T19:25:31.418Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/typed-answers-api-untapped-glance-qwen3-vl-4b/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "qqvtd6nwwdrgc0d1d8wth7hp60",
  "status": "succeeded",
  "output": {
    "answer": "Cat",
    "confidence": 1,
    "probabilities": [
      { "label": "Cat", "probability": 1 },
      { "label": "Dog", "probability": 0 },
      { "label": "Bird", "probability": 0 },
      { "label": "Horse", "probability": 0 }
    ]
  },
  "error": null,
  "created_at": "2026-10-23T19:25:31.418Z",
  "started_at": "2026-10-23T19:25:32.418Z",
  "completed_at": "2026-10-23T19:26:18.418Z"
}
```

`output` is a JSON object with `answer` (the chosen label), `confidence` (a 0 to 1 confidence value for that answer) and `probabilities` (a list of `{label, probability}` for every option).

### 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 `question` or a `question_type` that is not `yes_no` or `choice`.
- **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. Check that an image was sent and that `options_json` is a valid JSON array string. 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/typed-answers-api-untapped-glance-qwen3-vl-4b";

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: "65c82d4fbfad89eaee2dabd4abd0be8703d7a3187e216edbde26eecc6f56d69f", 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/glance-qwen3-vl-4b-input-api-market.jpg",
  "question": "Which animal is shown in this image?",
  "options_json": "[\"Cat\",\"Dog\",\"Bird\",\"Horse\"]",
  "question_type": "choice"
}).then((result) => console.log(result.answer, result.confidence));
```

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

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/typed-answers-api-untapped-glance-qwen3-vl-4b"


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": "65c82d4fbfad89eaee2dabd4abd0be8703d7a3187e216edbde26eecc6f56d69f", "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")


result = create_and_wait_for_prediction({
    "image": "https://blog.api.market/wp-content/uploads/2026/10/glance-qwen3-vl-4b-input-api-market.jpg",
    "question": "Which animal is shown in this image?",
    "options_json": "[\"Cat\",\"Dog\",\"Bird\",\"Horse\"]",
    "question_type": "choice"
})
print(result["answer"], result["confidence"])
```

## FAQ
- **When should I use `yes_no` and when `choice`?**
  - Use `yes_no` (the default) for questions with a yes or no answer. Use `choice` when you want the model to pick one of your labels, and list them in `options_json`.
- **Why is `options_json` a string?**
  - The field takes a JSON-encoded array. Build the list in your code and send it with `JSON.stringify(list)` or `json.dumps(list)`.
- **Can I send the image without hosting it?**
  - Yes. Put a base64-encoded image in `image_base64` instead of a URL in `image`.
- **Does it return a free-text explanation?**
  - No. It returns only the chosen answer, its confidence and the probability of each option.
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page. You can also email our support team at hello@magicapi.com.


Classify images with your own labels using Glance on Qwen3-VL-4B: yes/no or multiple-choice questions, JSON answers with per-option probabilities.