Name: aramintak Flux Softserve Anime Illustration Generator API
Brand: magicapi
SKU: api-magicapi-anime-img-generation-aramintak-flux-softserve-anime
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎨 Soft Anime Illustration Generator API (aramintak Flux Softserve Anime)

## Overview
Flux Softserve Anime is a LoRA style for the FLUX image model, published by aramintak. It generates soft, anime-style illustrations. Its developer says to use "sftsrv style illustration" in the prompt to trigger the style; the example below uses "sftsrv style".

Send a text `prompt` and the model returns a list of image URLs (1 to 4 images, set by `num_outputs`). You can also pass an `image` for image-to-image, or an `image` plus a `mask` for inpainting.

## Features
- **Soft anime illustration style**: triggered by "sftsrv style" in the prompt.
- **Text to image, image to image and inpainting**: use `prompt` alone, add `image` (with `prompt_strength`), or add `image` and `mask`.
- **Aspect ratios**: `1:1` (default), `16:9`, `21:9`, `3:2`, `2:3`, `4:5`, `5:4`, `3:4`, `4:3`, `9:16`, `9:21`, or `custom` with `width` and `height` (256 to 1440 px).
- **Two base models**: `dev` (default, about 28 steps) or `schnell` (about 4 steps).
- **Up to 4 images per request** with `num_outputs`.
- **Output formats**: `webp` (default), `jpg` or `png`, with `output_quality` 0 to 100.
- **Style strength**: `lora_scale` (-1 to 3, default 1), plus an optional second LoRA with `extra_lora` and `extra_lora_scale`.
- **Speed mode**: `go_fast` runs an fp8-quantized model for faster results.

## Use Cases
1. **Character and avatar art**: anime-style portraits for profiles, games and stories.
2. **Illustrated content**: soft anime visuals for blogs, newsletters and social posts.
3. **Concept sketches**: quick style explorations for scenes and characters.
4. **Restyling images**: send a photo or drawing as `image` and redraw it in the anime style.

## Visual Examples
### Example Input
Prompt: "a smart person, sftsrv style"

Settings: `model` `dev`, `aspect_ratio` `1:1`, `num_outputs` `1`, `num_inference_steps` `28`, `guidance_scale` `3.5`, `lora_scale` `1`, `output_format` `webp`, `output_quality` `80`.

### Example Output
![Flux Softserve Anime output for the prompt a smart person, a soft anime-style illustration](https://blog.api.market/wp-content/uploads/2026/10/flux-softserve-anime-output-421f81y1-api-market.webp)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | Yes |  |  | Prompt for the generated image. Include the trigger phrase "sftsrv style" to activate the trained style |
| `image` | string (URL) | No | none |  | Input image for image-to-image or inpainting mode. If provided, `aspect_ratio`, `width` and `height` are ignored |
| `mask` | string (URL) | No | none |  | Image mask for inpainting mode. If provided, `aspect_ratio`, `width` and `height` are ignored |
| `aspect_ratio` | string | No | `1:1` | `1:1`, `16:9`, `21:9`, `3:2`, `2:3`, `4:5`, `5:4`, `3:4`, `4:3`, `9:16`, `9:21`, `custom` | Aspect ratio for the generated image. `custom` uses `width` and `height` and runs in bf16 mode |
| `height` | integer | No | none | 256 to 1440 | Height of the generated image. Only works if `aspect_ratio` is `custom`. Rounded to the nearest multiple of 16. Incompatible with `go_fast` |
| `width` | integer | No | none | 256 to 1440 | Width of the generated image. Only works if `aspect_ratio` is `custom`. Rounded to the nearest multiple of 16. Incompatible with `go_fast` |
| `prompt_strength` | number | No | `0.8` | 0 to 1 | Prompt strength when using image-to-image. 1.0 fully replaces the information in the input image |
| `model` | string | No | `dev` | `dev`, `schnell` | Which model to run. `dev` performs best with around 28 inference steps; `schnell` only needs 4 |
| `num_outputs` | integer | No | `1` | 1 to 4 | Number of images to generate |
| `num_inference_steps` | integer | No | `28` | 1 to 50 | Number of denoising steps. More steps can give more detailed images, but take longer |
| `guidance_scale` | number | No | `3` | 0 to 10 | Guidance scale for the diffusion process. Lower values can give more realistic images. Good values to try are 2, 2.5, 3 and 3.5 |
| `seed` | integer | No | random |  | Random seed. Set it for reproducible generation |
| `output_format` | string | No | `webp` | `webp`, `jpg`, `png` | Format of the output images |
| `output_quality` | integer | No | `80` | 0 to 100 | Quality when saving the output images. 100 is best quality. Not relevant for `png` |
| `disable_safety_checker` | boolean | No | `false` |  | Disable the safety checker for generated images |
| `go_fast` | boolean | No | `false` |  | Run faster predictions with a model optimized for speed (currently fp8 quantized); disable to run in original bf16 |
| `megapixels` | string | No | `1` | `1`, `0.25` | Approximate number of megapixels for the generated image |
| `lora_scale` | number | No | `1` | -1 to 3 | How strongly the main LoRA is applied. Sane results between 0 and 1 for base inference. With `go_fast` a 1.5x multiplier is applied to this value |
| `extra_lora` | string | No | none |  | Extra LoRA weights to load: a Hugging Face URL (`huggingface.co/<owner>/<model-name>`), a CivitAI URL (`civitai.com/models/<id>[/<model-name>]`) or a direct `.safetensors` URL |
| `extra_lora_scale` | number | No | `1` | -1 to 3 | How strongly the extra LoRA is applied. Sane results between 0 and 1 for base inference. With `go_fast` a 1.5x multiplier is applied to this value |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.05 per API unit.
- **1 API unit = 1 request with up to 26 seconds of processing.** A request that runs longer is billed one more unit for each started 26-second block (for example, more than 26 and up to 52 seconds of processing is 2 units, $0.10).
- Processing time grows with `num_outputs`, `num_inference_steps` and image size; `schnell` and `go_fast` are faster.
- **Example cost**: the example request (one image, 28 steps, `dev`) took 15.1 seconds of processing, which fits in one 26-second block: 1 API unit, $0.05.
- **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/anime-img-generation-aramintak-flux-softserve-anime`

### 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/anime-img-generation-aramintak-flux-softserve-anime/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "9e35b00131765c22d260dce4106d6688e83e67d4416955ca137cb27c94ed81c9",
    "input": {
      "model": "dev",
      "prompt": "a smart person, sftsrv style",
      "lora_scale": 1,
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 3.5,
      "output_quality": 80,
      "num_inference_steps": 28
    }
  }'
```

Response (201 Created):
```json
{
  "id": "421f81y1rxrmy0d1f19sb4qtm4",
  "version": "9e35b00131765c22d260dce4106d6688e83e67d4416955ca137cb27c94ed81c9",
  "status": "starting",
  "created_at": "2026-10-26T13:08:37.598Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/anime-img-generation-aramintak-flux-softserve-anime/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "421f81y1rxrmy0d1f19sb4qtm4",
  "status": "succeeded",
  "output": [
    "https://blog.api.market/wp-content/uploads/2026/10/flux-softserve-anime-output-421f81y1-api-market.webp"
  ],
  "error": null,
  "created_at": "2026-10-26T13:08:37.598Z",
  "started_at": "2026-10-26T13:08:37.934Z",
  "completed_at": "2026-10-26T13:08:53.006Z"
}
```

`output` is an array of image URLs, one per generated image (`num_outputs`).

### 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`, a `num_outputs` above 4 or an `aspect_ratio` that is not in the list.
- **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`; 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/anime-img-generation-aramintak-flux-softserve-anime";

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: "9e35b00131765c22d260dce4106d6688e83e67d4416955ca137cb27c94ed81c9", 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({
  "model": "dev",
  "prompt": "a smart person, sftsrv style",
  "lora_scale": 1,
  "num_outputs": 1,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 3.5,
  "output_quality": 80,
  "num_inference_steps": 28
}).then((urls) => console.log(urls));
```

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

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/anime-img-generation-aramintak-flux-softserve-anime"


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": "9e35b00131765c22d260dce4106d6688e83e67d4416955ca137cb27c94ed81c9", "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({
    "model": "dev",
    "prompt": "a smart person, sftsrv style",
    "lora_scale": 1,
    "num_outputs": 1,
    "aspect_ratio": "1:1",
    "output_format": "webp",
    "guidance_scale": 3.5,
    "output_quality": 80,
    "num_inference_steps": 28
}))
```

## FAQ
- **Do I need the trigger phrase?**
  - Yes, for the soft anime look. Include "sftsrv style" (or "sftsrv style illustration") in the prompt; without it the trained style is less likely to appear.
- **How many images does one request return?**
  - Between 1 and 4, set by `num_outputs` (default 1). Billing is by processing time, so more images take longer and can add units.
- **How do I make generation faster and cheaper?**
  - Use `model` `schnell` with about 4 `num_inference_steps`, turn on `go_fast`, or set `megapixels` to `0.25`. Shorter processing means fewer 26-second units.
- **Can I turn a photo into this style?**
  - Yes. Pass it as `image` and lower `prompt_strength` to keep more of the original. Add a `mask` to repaint only part of it.
- **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 or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


Create soft anime-style illustrations with the Flux Softserve Anime LoRA: text-to-image, image-to-image and inpainting. $0.05 per request, pay as you go.