Name: GPT-4o mini API: Fast, Low-Cost LLM by OpenAI
Brand: magicapi
SKU: api-magicapi-low-latency-ai-language-model-openai-gpt-4o-mini
Availability: InStock
Author: API Market

Shashank Agarwal

# ⚡ Fast Language Model API (OpenAI GPT-4o mini)

## Overview
GPT-4o mini is a language model from OpenAI. Its developer describes it as a low-latency, low-cost version of GPT-4o.

Send a `prompt` with an optional `system_prompt`, or a full chat history in `messages`. Add image URLs in `image_input` to ask about images. The reply comes back as a list of text chunks; join them in order to get the full text. You pay per token: input and output tokens are counted separately.

## Features
- **Prompt or chat input**: send one `prompt` or a `messages` list. When `messages` is set, `prompt` and `system_prompt` are ignored.
- **Image input**: `image_input` takes a list of image URLs for the model to read.
- **System prompt**: set the assistant's role and rules with `system_prompt`.
- **Length control**: `max_completion_tokens` caps the output (default 4096).
- **Sampling controls**: `temperature` (0 to 2, default 1), `top_p` (0 to 1, default 1), `frequency_penalty` and `presence_penalty` (-2 to 2, default 0).
- **Low token price**: $0.25 per million input tokens and $1 per million output tokens.

## Use Cases
1. **Chat assistants and support bots**: multi-turn conversations with `messages`.
2. **High-volume text tasks**: classification, extraction, summaries and rewrites where cost and latency matter.
3. **Image questions**: describe screenshots, read charts or compare product photos sent in `image_input`.
4. **Advice and writing help**: drafts, emails and step-by-step answers shaped by a `system_prompt`.

## Visual Examples
### Example Input
Prompt: "My situationship is starting to cross over into more than \"friends with benefits\" territory. I'm catching feelings, but this is not in my five year plan. What should I do?"

System prompt: "You are a helpful assistant." Sent with `max_completion_tokens` `4096`, `temperature` `1`, `top_p` `1`, both penalties at `0`, an empty `messages` list and no images.

### Example Output
The output chunks joined in order (54 input tokens, 395 output tokens):

```text
Navigating a situationship that’s shifting toward something deeper can be challenging, especially if it's not what you had in mind for your future. Here are some steps you can consider:

1. **Reflect on Your Feelings**: Take some time to understand your feelings. Are you developing feelings because of the connection you share, or is it more about the situation itself? Knowing what you're feeling and why can help clarify your next steps.

2. **Assess Your Goals**: Consider what you want in the long term. If having a serious relationship is not part of your five-year plan, think about whether you want to adjust your plan or if you prefer to maintain your current lifestyle.

3. **Communicate**: Talk to your situationship partner about how you feel. Open communication can help both of you understand where you stand. They may feel similarly, or they might not be ready for more, which could impact your decision.

4. **Set Boundaries**: If you decide you want to keep it casual, it might be helpful to establish clear boundaries to maintain that dynamic. This might involve limiting emotional intimacy or physical closeness to keep feelings in check.

5. **Consider Taking a Step Back**: If you're finding it hard to manage your feelings, you might need to create some distance. This can help you regain perspective on what you really want without the immediate emotional pull.

6. **Focus on Your Goals**: Concentrate on your five-year plan and goals. Engaging in activities and relationships that align with those goals can help you maintain your focus and steer your life in the direction you desire.

7. **Seek Support**: Sometimes, talking with friends or a trusted person about your feelings can provide insight and clarity that you might not see on your own.

Ultimately, listen to your intuition and prioritize what feels right for you. Balancing personal feelings with your future plans is important, and taking thoughtful steps can help you navigate this complex situation.
```

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | No | none |  | The prompt to send to the model. Do not use together with messages |
| `system_prompt` | string | No | none |  | System prompt that sets the assistant's behavior |
| `messages` | array of objects | No | `[]` |  | A list of messages, for example [{"role": "user", "content": "Hello, how are you?"}]. If provided, prompt and system_prompt are ignored |
| `image_input` | array of URLs | No | `[]` |  | List of image URLs to send to the model |
| `temperature` | number | No | `1` | `0` to `2` | Sampling temperature between 0 and 2 |
| `max_completion_tokens` | integer | No | `4096` |  | Maximum number of completion tokens to generate |
| `top_p` | number | No | `1` | `0` to `1` | Nucleus sampling: the model considers the tokens within the top_p probability mass (0.1 means only the top 10%) |
| `frequency_penalty` | number | No | `0` | `-2` to `2` | Positive values penalize tokens that have already appeared often, reducing repetition |
| `presence_penalty` | number | No | `0` | `-2` to `2` | Positive values penalize tokens that have already appeared at all, making the model more likely to move to new topics |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.001 per API unit.
- **1 API unit = 1,000 weighted tokens; an output token counts 1, an input token counts 0.25.**
- That works out to **$0.25 per million input tokens** and **$1 per million output tokens**.
- Each request is rounded up to a whole API unit, so every successful request costs at least $0.001. Images in `image_input` count as input tokens.
- **Example cost**: the example request used 54 input tokens and 395 output tokens: 54 x 0.25 + 395 = 408.5 weighted tokens, rounded up to 1 API unit = **$0.001**.
- **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/low-latency-ai-language-model-openai-gpt-4o-mini`

### 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/low-latency-ai-language-model-openai-gpt-4o-mini/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "openai/gpt-4o-mini",
    "input": {
      "top_p": 1,
      "prompt": "My situationship is starting to cross over into more than \"friends with benefits\" territory. I'\''m catching feelings, but this is not in my five year plan. What should I do?",
      "messages": [],
      "image_input": [],
      "temperature": 1,
      "system_prompt": "You are a helpful assistant.",
      "presence_penalty": 0,
      "frequency_penalty": 0,
      "max_completion_tokens": 4096
    }
  }'
```

Response (201 Created):
```json
{
  "id": "y9a0jygsfhrmw0d1c6tbq5v8kg",
  "version": "openai/gpt-4o-mini",
  "status": "starting",
  "created_at": "2026-10-22T03:42:35.557Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/low-latency-ai-language-model-openai-gpt-4o-mini/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "y9a0jygsfhrmw0d1c6tbq5v8kg",
  "status": "succeeded",
  "output": [
    "Navigating a situationship that’s shifting toward something deeper can be challenging, especially if it's not what you had in mind for your future. Here are some steps you can consider:\n\n1. **Reflect on Your Feelings**: Take some time to understand your feelings. Are you developing feelings because of the connection you share, or is it more about the situation itself? Knowing what you're feeling and why can help clarify your next steps.\n\n2. **Assess Your Goals**: Consider what you want in the long term. If having a serious relationship is not part of your five-year plan, think about whether you want to adjust your plan or if you prefer to maintain your current lifestyle.\n\n3. **Communicate**: Talk to your situationship partner about how you feel. Open communication can help both of you understand where you stand. They may feel similarly, or they might not be ready for more, which could impact your decision.\n\n4. **Set Boundaries**: If you decide you want to keep it casual, it might be helpful to establish clear boundaries to maintain that dynamic. This might involve limiting emotional intimacy or physical closeness to keep feelings in check.\n\n5. **Consider Taking a Step Back**: If you're finding it hard to manage your feelings, you might need to create some distance. This can help you regain perspective on what you really want without the immediate emotional pull.\n\n6. **Focus on Your Goals**: Concentrate on your five-year plan and goals. Engaging in activities and relationships that align with those goals can help you maintain your focus and steer your life in the direction you desire.\n\n7. **Seek Support**: Sometimes, talking with friends or a trusted person about your feelings can provide insight and clarity that you might not see on your own.\n\nUltimately, listen to your intuition and prioritize what feels right for you. Balancing personal feelings with your future plans is important, and taking thoughtful steps can help you navigate this complex situation."
  ],
  "error": null,
  "created_at": "2026-10-22T03:42:35.557Z",
  "started_at": "2026-10-22T03:42:36.557Z",
  "completed_at": "2026-10-22T03:42:41.977Z"
}
```

`output` is a list of text chunks. Join them in order to get the full reply. The example above shows the reply joined into one element; a real response usually splits it into many short chunks.

### 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 `temperature` above 2 or a `top_p` above 1.
- **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/low-latency-ai-language-model-openai-gpt-4o-mini";

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: "openai/gpt-4o-mini", 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({
  "top_p": 1,
  "prompt": "My situationship is starting to cross over into more than \"friends with benefits\" territory. I'm catching feelings, but this is not in my five year plan. What should I do?",
  "messages": [],
  "image_input": [],
  "temperature": 1,
  "system_prompt": "You are a helpful assistant.",
  "presence_penalty": 0,
  "frequency_penalty": 0,
  "max_completion_tokens": 4096
}).then((output) => console.log(output.join("")));
```

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

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/low-latency-ai-language-model-openai-gpt-4o-mini"


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": "openai/gpt-4o-mini", "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({
    "top_p": 1,
    "prompt": "My situationship is starting to cross over into more than \"friends with benefits\" territory. I'm catching feelings, but this is not in my five year plan. What should I do?",
    "messages": [],
    "image_input": [],
    "temperature": 1,
    "system_prompt": "You are a helpful assistant.",
    "presence_penalty": 0,
    "frequency_penalty": 0,
    "max_completion_tokens": 4096
})
print("".join(output))
```

## FAQ
- **How do I get the full reply text?**
  - `output` is an array of text chunks. Join them in order with no separator: `output.join("")` in JavaScript or `"".join(output)` in Python.
- **Should I use prompt or messages?**
  - Use `prompt` (with an optional `system_prompt`) for a single question. Use `messages` for a conversation; when `messages` is set, `prompt` and `system_prompt` are ignored.
- **Can I send more than one image?**
  - Yes. `image_input` is a list of image URLs. Images count as input tokens.
- **How much does a request cost?**
  - $0.25 per million input tokens and $1 per million output tokens, rounded up to a whole API unit ($0.001) per request. The example request cost $0.001.
- **Why did the reply stop early?**
  - The model stops when it reaches `max_completion_tokens` (default 4096). Raise it for longer answers.
- **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, 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.


Call OpenAI GPT-4o mini for chat, writing and image questions with a prompt, chat messages or image URLs. Pay as you go, no monthly fee.