Name: IBM Granite 4.0 H Small API: Long-Context Instruct LLM
Brand: magicapi
SKU: api-magicapi-long-context-instruct-ai-ibm-granite-granite-4-0-h-small
Availability: InStock
Author: API Market

Shashank Agarwal

# 💬 Long-Context Instruct LLM API (IBM Granite 4.0 H Small)

## Overview
Granite 4.0 H Small is a 32B-parameter long-context instruct model from IBM, fine-tuned from Granite 4.0 H Small Base on open-source instruction datasets with permissive licenses and on IBM's own synthetic datasets.

Send a single `prompt` (with an optional `system_prompt`) or a `messages` array for chat. You can also pass `documents` and `tools` to the chat template and constrain the answer with `response_format`. The model returns generated text.

## Features
- **32B parameters, long context**: the largest model in this group of IBM Granite listings.
- **Prompt or chat**: send a single `prompt` (with optional `system_prompt`) or a `messages` array for multi-turn chat.
- **Documents and tools**: `documents`, `tools` and `tool_choice` are passed to the model's chat template for grounded answers and function calling.
- **Structured output**: `response_format` specifies the format the model must output.
- **Length controls**: `max_completion_tokens` caps the generated tokens; `min_tokens` sets a minimum. `max_tokens` is deprecated.
- **Sampling controls**: `temperature`, `top_p`, `top_k`, `presence_penalty`, `frequency_penalty`, `repetition_penalty`, `stop` and `seed`.

## Use Cases
1. **Long-document work**: summaries and Q&A over long reports passed in `prompt` or `documents`.
2. **Retrieval-augmented answers**: pass retrieved passages in `documents` and ask about them.
3. **Function calling**: describe your functions in `tools` and let the model choose with `tool_choice`.
4. **Structured extraction**: request a fixed output shape with `response_format`.

## Visual Examples
### Example Input
Prompt: "What is Pi?"

Settings: `temperature` 0.6, `top_p` 0.9, `top_k` 50, `min_tokens` 0, `presence_penalty` 0, `frequency_penalty` 0, `stream` `false`, `add_generation_prompt` `true`; `messages`, `documents`, `tools` and `stop` empty.

### Example Output
The chunks of this run joined into one answer:

````markdown
Pi (π) is a mathematical constant that represents the ratio of a circle's circumference to its diameter. It is approximately equal to 3.14159, although it is an irrational number, meaning it cannot be exactly expressed as a simple fraction and its decimal representation never ends or repeats. Pi is used in various areas of mathematics, including geometry, trigonometry, and calculus, as well as in physics and engineering.
````

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | No | none |  | Completion API user prompt. |
| `messages` | array of objects | No | `[]` |  | Chat messages, as message objects (for example `{"role": "user", "content": "..."}`). Use instead of `prompt` for multi-turn chat. |
| `documents` | array of objects | No | `[]` |  | Documents for the request. Passed to the chat template. |
| `tools` | array of objects | No | `[]` |  | Tools for the request. Passed to the chat template. |
| `tool_choice` | string | No | none |  | Tool choice for the request. If the choice is a specific function, pass it as a JSON string. |
| `response_format` | object | No | none |  | An object specifying the format that the model must output. |
| `system_prompt` | string | No | none |  | Completion API system prompt. The chat template provides a good default. |
| `chat_template` | string | No | none |  | A template to format the prompt with. If not specified, the chat template provided by the model will be used. |
| `add_generation_prompt` | boolean | No | `true` |  | Add generation prompt. Passed to the chat template. |
| `chat_template_kwargs` | object | No | `{}` |  | Additional arguments to be passed to the chat template. |
| `min_tokens` | integer | No | `0` |  | The minimum number of tokens the model should generate as output. |
| `max_tokens` | integer | No | none |  | Deprecated in favor of `max_completion_tokens`. |
| `max_completion_tokens` | integer | No | none |  | An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens. |
| `temperature` | number | No | `0` |  | The value used to modulate the next token probabilities. |
| `top_p` | number | No | `0.9` |  | A probability threshold for generating the output. If < 1.0, only keep the top tokens with cumulative probability >= top_p (nucleus filtering). |
| `top_k` | integer | No | `50` |  | The number of highest probability tokens to consider for generating the output. If > 0, only keep the top k tokens with highest probability (top-k filtering). |
| `presence_penalty` | number | No | none |  | Presence penalty |
| `frequency_penalty` | number | No | none |  | Frequency penalty |
| `repetition_penalty` | number | No | none |  | Repetition penalty |
| `stop` | array of strings | No | `[]` |  | A list of sequences to stop generation at. For example, `["<end>","<stop>"]` stops generation at the first instance of `<end>` or `<stop>`. |
| `seed` | integer | No | none |  | Random seed. Leave unspecified to randomize the seed. |
| `stream` | boolean | No | `false` |  | Request a streaming response. Results are read by polling GET `/predictions/{id}`, so leave this `false`. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.0032 per API unit.
- **1 API unit = 10,000 weighted tokens**; an output token counts 1, an input token counts 0.24.
- That is **$0.0768 per million input tokens** and **$0.32 per million output tokens**.
- Each request is rounded up to a whole API unit, so the smallest charge per request is $0.0032.
- **Example cost**: the example request used 34 input tokens and 86 output tokens. 34 x 0.24 + 86 = 94.16 weighted tokens, rounded up to 1 API unit = **$0.0032**.
- **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/long-context-instruct-ai-ibm-granite-granite-4-0-h-small`

### 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/long-context-instruct-ai-ibm-granite-granite-4-0-h-small/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "ibm-granite/granite-4.0-h-small",
    "input": {
      "stop": [],
      "tools": [],
      "top_k": 50,
      "top_p": 0.9,
      "prompt": "What is Pi?",
      "stream": false,
      "messages": [],
      "documents": [],
      "min_tokens": 0,
      "temperature": 0.6,
      "presence_penalty": 0,
      "frequency_penalty": 0,
      "chat_template_kwargs": {},
      "add_generation_prompt": true
    }
  }'
```

Response (201 Created):
```json
{
  "id": "8a4t76nnk9rn80d1d0zb6kyczc",
  "version": "ibm-granite/granite-4.0-h-small",
  "status": "starting",
  "created_at": "2026-10-23T10:11:42.830Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/long-context-instruct-ai-ibm-granite-granite-4-0-h-small/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "8a4t76nnk9rn80d1d0zb6kyczc",
  "status": "succeeded",
  "output": [
    "Pi (π) is a mathematical constant that represents the ratio of a circle's circumference to its diameter. It is approximately equal to 3.14159, although it is an irrational number, meaning it cannot be exactly expressed as a simple fraction and its decimal representation never ends or repeats. Pi is used in various areas of mathematics, including geometry, trigonometry, and calculus, as well as in physics and engineering."
  ],
  "error": null,
  "created_at": "2026-10-23T10:11:42.830Z",
  "started_at": "2026-10-23T10:14:27.392Z",
  "completed_at": "2026-10-23T10:14:28.887Z"
}
```

`output` is an array of text chunks. Join them in order to get the full answer. The example above shows the chunks from this run already joined into one string.

### 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 `messages` that is not an array.
- **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/long-context-instruct-ai-ibm-granite-granite-4-0-h-small";

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: "ibm-granite/granite-4.0-h-small", 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.join("");
    }

    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({
    "stop": [],
    "tools": [],
    "top_k": 50,
    "top_p": 0.9,
    "prompt": "What is Pi?",
    "stream": false,
    "messages": [],
    "documents": [],
    "min_tokens": 0,
    "temperature": 0.6,
    "presence_penalty": 0,
    "frequency_penalty": 0,
    "chat_template_kwargs": {},
    "add_generation_prompt": true
  }).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/long-context-instruct-ai-ibm-granite-granite-4-0-h-small"


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": "ibm-granite/granite-4.0-h-small", "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 "".join(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({
    "stop": [],
    "tools": [],
    "top_k": 50,
    "top_p": 0.9,
    "prompt": "What is Pi?",
    "stream": False,
    "messages": [],
    "documents": [],
    "min_tokens": 0,
    "temperature": 0.6,
    "presence_penalty": 0,
    "frequency_penalty": 0,
    "chat_template_kwargs": {},
    "add_generation_prompt": True
})
print(output)
```

## FAQ
- **Why does `output` come back as a list?**
  - The answer is returned as an array of text chunks. Join them in order (`output.join("")` in JavaScript, `"".join(output)` in Python).
- **Should I use `prompt` or `messages`?**
  - Use `prompt` (and optionally `system_prompt`) for a single question. Use `messages` to send a whole conversation as message objects with a `role` and `content`.
- **How do I limit the answer length?**
  - Set `max_completion_tokens`. `max_tokens` still exists but is deprecated in favor of `max_completion_tokens`.
- **How is a request billed?**
  - By weighted tokens at $0.0032 per 10,000: input tokens count 0.24 and output tokens count 1, rounded up to a whole unit per request. Most short requests cost $0.0032.
- **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.


Run IBM Granite 4.0 H Small, a 32B long-context instruct model with chat messages, documents, tools and structured output. Pay as you go, no monthly fee.