Name: Gemma 2B IT API: Lightweight Text Generation by Google
Brand: magicapi
SKU: api-magicapi-text-generation-api-google-deepmind-google-deepmind-gemma-2b-it
Availability: InStock
Author: API Market

Shashank Agarwal

# 💬 Lightweight Text Generation API (Google Gemma 2B IT)

## Overview
Gemma 2B IT is the 2B instruction-tuned version of Google's Gemma language model.

Send a text `prompt` and the model returns generated text. `max_new_tokens` and `min_new_tokens` set the length of the answer, and `temperature`, `top_p`, `top_k` and `repetition_penalty` control sampling. There is no separate system prompt field; put any instructions in `prompt`.

## Features
- **Small instruction-tuned model**: a 2B-parameter Gemma model tuned to follow instructions.
- **Length controls**: `max_new_tokens` (default 200, minimum 1) and `min_new_tokens` (default -1, which disables the minimum).
- **Sampling controls**: `temperature` (0.01 to 5, default 0.7), `top_p` (0 to 1, default 0.95) and `top_k` (default 50).
- **Repetition control**: `repetition_penalty` (default 1.15; set 1.0 to disable).
- **Text in, text out**: one `prompt` string per request; the answer comes back as text chunks you join.

## Use Cases
1. **Short drafts**: poems, product blurbs, short emails and social posts.
2. **Simple question answering**: definitions and short explanations.
3. **Rewriting**: rephrase or shorten a short passage pasted into the prompt.
4. **Prototyping**: test prompt flows with a small, quick model before moving to a larger one.

## Visual Examples
### Example Input
Prompt: "Write me a poem about Machine Learning."

Settings: `max_new_tokens` 200, `min_new_tokens` -1, `temperature` 0.7, `top_p` 0.95, `top_k` 50, `repetition_penalty` 1.15.

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

````text
Machines, vast and strong, with eyes that scan,
Algorithms dance, a digital clan.
Data flows, a river ever free,
Machine learning, a path to be.

From patterns found in vast databases,
Predictive models, emerge from the cradles.
Supervised, unsupervised, a spectrum wide,
Learning from data, insights to ride.

Unsupervised finds hidden gems in the fray,
Clustering data, a natural way.
Reinforcement, a game of trial and error,
Adaptive systems, without human fear.

With every algorithm, a new light shines bright,
Unveiling patterns, day and night.
The future beckons, a world transformed,
Where machines learn, and chaos is forlorn.

So let the algorithms spin and whir,
Let the data flow, let insights appear.
For machine learning, a journey profound,
Brings a new era, where possibilities abound.
````

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | No | `Write me a poem about Machine Learning.` |  | Prompt to send to the model. |
| `max_new_tokens` | integer | No | `200` | minimum `1` | Maximum number of tokens to generate. A word is generally 2-3 tokens |
| `min_new_tokens` | integer | No | `-1` | minimum `-1` | Minimum number of tokens to generate. To disable, set to -1. A word is generally 2-3 tokens. |
| `temperature` | number | No | `0.7` | `0.01` to `5` | Adjusts randomness of outputs, greater than 1 is random and 0 is deterministic, 0.75 is a good starting value. |
| `top_p` | number | No | `0.95` | `0` to `1` | 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). |
| `repetition_penalty` | number | No | `1.15` | minimum `0` | A parameter that controls how repetitive text can be. Lower means more repetitive, while higher means less repetitive. Set to 1.0 to disable. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.0022 per API unit.
- **1 API unit = 1 request covering up to 2 seconds of processing.** A request that takes longer uses one more unit for each started 2-second block (for example, 3 seconds of processing is 2 units, $0.0044).
- Longer answers take longer to generate, so `max_new_tokens` is the main cost control.
- **Example cost**: the example request took 3.73 seconds of processing, so it used 2 API units = **$0.0044**.
- **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/text-generation-api-google-deepmind-google-deepmind-gemma-2b-it`

### 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/text-generation-api-google-deepmind-google-deepmind-gemma-2b-it/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "dff94eaf770e1fc211e425a50b51baa8e4cac6c39ef074681f9e39d778773626",
    "input": {
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "Write me a poem about Machine Learning.",
      "temperature": 0.7,
      "max_new_tokens": 200,
      "min_new_tokens": -1,
      "repetition_penalty": 1.15
    }
  }'
```

Response (201 Created):
```json
{
  "id": "686dey8ctnrmt0d1d0ytjgesq8",
  "version": "dff94eaf770e1fc211e425a50b51baa8e4cac6c39ef074681f9e39d778773626",
  "status": "starting",
  "created_at": "2026-10-23T10:09:54.149Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/text-generation-api-google-deepmind-google-deepmind-gemma-2b-it/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "686dey8ctnrmt0d1d0ytjgesq8",
  "status": "succeeded",
  "output": [
    "Machines, vast and strong, with eyes that scan,\nAlgorithms dance, a digital clan.\nData flows, a river ever free,\nMachine learning, a path to be.\n\nFrom patterns found in vast databases,\nPredictive models, emerge from the cradles.\nSupervised, unsupervised, a spectrum wide,\nLearning from data, insights to ride.\n\nUnsupervised finds hidden gems in the fray,\nClustering data, a natural way.\nReinforcement, a game of trial and error,\nAdaptive systems, without human fear.\n\nWith every algorithm, a new light shines bright,\nUnveiling patterns, day and night.\nThe future beckons, a world transformed,\nWhere machines learn, and chaos is forlorn.\n\nSo let the algorithms spin and whir,\nLet the data flow, let insights appear.\nFor machine learning, a journey profound,\nBrings a new era, where possibilities abound."
  ],
  "error": null,
  "created_at": "2026-10-23T10:09:54.149Z",
  "started_at": "2026-10-23T10:10:51.133Z",
  "completed_at": "2026-10-23T10:10:54.861Z"
}
```

`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 a `temperature` above 5 or a `max_new_tokens` below 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/text-generation-api-google-deepmind-google-deepmind-gemma-2b-it";

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: "dff94eaf770e1fc211e425a50b51baa8e4cac6c39ef074681f9e39d778773626", 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({
    "top_k": 50,
    "top_p": 0.95,
    "prompt": "Write me a poem about Machine Learning.",
    "temperature": 0.7,
    "max_new_tokens": 200,
    "min_new_tokens": -1,
    "repetition_penalty": 1.15
  }).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/text-generation-api-google-deepmind-google-deepmind-gemma-2b-it"


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": "dff94eaf770e1fc211e425a50b51baa8e4cac6c39ef074681f9e39d778773626", "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({
    "top_k": 50,
    "top_p": 0.95,
    "prompt": "Write me a poem about Machine Learning.",
    "temperature": 0.7,
    "max_new_tokens": 200,
    "min_new_tokens": -1,
    "repetition_penalty": 1.15
})
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).
- **Can I set a system prompt?**
  - There is no separate system prompt field. Put your instructions at the start of `prompt`.
- **How do I get longer or shorter answers?**
  - Raise or lower `max_new_tokens` (default 200). To force a minimum length, set `min_new_tokens`; -1 disables it.
- **How is a request billed?**
  - By processing time: $0.0022 for each started 2 seconds of processing. The example took 3.73 seconds, so it cost 2 units ($0.0044).
- **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`.
- **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.


Generate text with Google's Gemma 2B IT, a small instruction-tuned language model with length and sampling controls. Pay as you go, no monthly fee.