Name: Pruna AI P-Image-LoRA Text-to-Image with LoRA API
Brand: magicapi
SKU: api-magicapi-text-to-image-lora-prunaai-p-image-lora
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎨 Text-to-Image with LoRA API (Pruna AI P-Image-LoRA)

## Overview
P-Image-LoRA is a text-to-image model from Pruna AI that can load a trained LoRA (a small set of add-on weights) to generate images in a specific style or with a specific subject. You describe the image in `prompt` and point `lora_weights` at a LoRA on HuggingFace.

Each request returns the URL of one image. `lora_weights` is optional: without it the model generates from the prompt alone.

## Features
- **LoRA styles**: load LoRA weights from a HuggingFace URL in the format `huggingface.co/<owner>/<model-name>[/<lora-weights-file.safetensors>]`.
- **LoRA strength**: `lora_scale` (-1 to 3, default `0.5`) sets how strongly the LoRA is applied; its developer notes that 0.5 usually works well for most LoRAs.
- **Aspect ratios**: `1:1`, `16:9` (default), `9:16`, `4:3`, `3:4`, `3:2`, `2:3`, or `custom`.
- **Custom size**: with `aspect_ratio` set to `custom`, set `width` and `height` (256 to 1440, multiples of 16).
- **Prompt upsampling**: `prompt_upsampling` (default `false`) rewrites the prompt with a language model before generation.
- **Seed**: set `seed` to reproduce a result.

## Use Cases
1. **Consistent visual style**: generate a series of images that all share the look of one style LoRA, such as pixel art.
2. **Game and app art**: produce icons, scenes and backgrounds in a fixed art style.
3. **Brand or character LoRAs**: generate new scenes with a subject or style you trained a LoRA on.
4. **Test your own LoRAs**: try LoRAs trained for this model with different prompts and strengths.

## Visual Examples
### Example Input
Prompt: "pixel art style A cozy cabin in a snowy forest with smoke from the chimney."

LoRA: `https://huggingface.co/davidberenstein1957/p-image-pixel-art-lora/resolve/main/weights.safetensors` with `lora_scale` `0.5`, `aspect_ratio` `1:1`, `prompt_upsampling` `false`.

### Example Output
![P-Image-LoRA output: a pixel art cabin in a snowy forest with smoke rising from the chimney](https://blog.api.market/wp-content/uploads/2026/11/p-image-lora-output-s9yck1wc-api-market.jpeg)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | Yes |  |  | Text prompt for image generation |
| `aspect_ratio` | string | No | `16:9` | `1:1`, `16:9`, `9:16`, `4:3`, `3:4`, `3:2`, `2:3`, `custom` | Aspect ratio for the generated image |
| `width` | integer | No | none | 256 to 1440 | Width of the generated image. Only used when `aspect_ratio` is `custom`. Must be a multiple of 16 |
| `height` | integer | No | none | 256 to 1440 | Height of the generated image. Only used when `aspect_ratio` is `custom`. Must be a multiple of 16 |
| `prompt_upsampling` | boolean | No | `false` |  | Upsample (rewrite) the prompt with a language model |
| `seed` | integer | No | none |  | Random seed. Set it for reproducible generation |
| `disable_safety_checker` | boolean | No | `false` |  | Disable the safety checker for generated images |
| `lora_weights` | string | No | none |  | LoRA weights to load. Supports HuggingFace URLs in the format huggingface.co/&lt;owner&gt;/&lt;model-name&gt;[/&lt;lora-weights-file.safetensors&gt;] |
| `lora_scale` | number | No | `0.5` | -1 to 3 | How strongly the LoRA is applied. 0.5 usually works well for most LoRAs |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.0059 per API unit.
- **1 API unit = 1 output image.** Each request returns one image (no input changes the count), so each successful request costs $0.0059.
- The example request above returned 1 image and cost $0.0059.
- **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-to-image-lora-prunaai-p-image-lora`

### 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-to-image-lora-prunaai-p-image-lora/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "prunaai/p-image-lora",
    "input": {
      "prompt": "pixel art style A cozy cabin in a snowy forest with smoke from the chimney.",
      "lora_scale": 0.5,
      "aspect_ratio": "1:1",
      "lora_weights": "https://huggingface.co/davidberenstein1957/p-image-pixel-art-lora/resolve/main/weights.safetensors",
      "prompt_upsampling": false
    }
  }'
```

Response (201 Created):
```json
{
  "id": "s9yck1wcvsrmw0d1mvnsfjstbr",
  "version": "prunaai/p-image-lora",
  "status": "starting",
  "created_at": "2026-11-04T14:16:57.310Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/text-to-image-lora-prunaai-p-image-lora/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "s9yck1wcvsrmw0d1mvnsfjstbr",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/11/p-image-lora-output-s9yck1wc-api-market.jpeg",
  "error": null,
  "created_at": "2026-11-04T14:16:57.310Z",
  "started_at": "2026-11-04T14:16:57.322Z",
  "completed_at": "2026-11-04T14:17:01.696Z"
}
```

`output` is a single URL string pointing to the generated image.

### 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 `prompt`, an `aspect_ratio` that is not in the list, or a `width` outside 256 to 1440.
- **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
Requests are rate limited to keep usage fair. If you receive a 429 response, wait and retry with exponential backoff.

## 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`, for example when the LoRA URL cannot be loaded; 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-to-image-lora-prunaai-p-image-lora";

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: "prunaai/p-image-lora", 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({
  prompt: "pixel art style A cozy cabin in a snowy forest with smoke from the chimney.",
  lora_scale: 0.5,
  aspect_ratio: "1:1",
  lora_weights: "https://huggingface.co/davidberenstein1957/p-image-pixel-art-lora/resolve/main/weights.safetensors",
  prompt_upsampling: false
}).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-to-image-lora-prunaai-p-image-lora"


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": "prunaai/p-image-lora", "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")


print(create_and_wait_for_prediction({
    "prompt": "pixel art style A cozy cabin in a snowy forest with smoke from the chimney.",
    "lora_scale": 0.5,
    "aspect_ratio": "1:1",
    "lora_weights": "https://huggingface.co/davidberenstein1957/p-image-pixel-art-lora/resolve/main/weights.safetensors",
    "prompt_upsampling": False
}))
```

## FAQ
- **Which LoRAs can I use?**
  - LoRAs trained for P-Image, loaded from a public HuggingFace URL in `lora_weights`. Point to the `.safetensors` file or to the model repository.
- **Do I have to send a LoRA?**
  - No. `lora_weights` is optional; without it the model generates from the prompt alone.
- **How do I get an exact image size?**
  - Set `aspect_ratio` to `custom` and send `width` and `height` between 256 and 1440, each a multiple of 16. They are ignored for the other aspect ratios.
- **Does a LoRA need a trigger word in the prompt?**
  - Many style LoRAs respond to a phrase they were trained on, such as "pixel art style" in the example. Check the LoRA's page for its trigger phrase.
- **How many images does one request return?**
  - One. Each image is billed as one API unit ($0.0059).
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, check that the LoRA URL is reachable, 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.


Generate images from text with Pruna AI's P-Image-LoRA, optionally styled by a trained LoRA from HuggingFace. $0.0059 per image, pay as you go.