---
title: Step 3: Generate Your First Image (Text-to-Image) | Scenario Docs
---

Now that you have authenticated your requests, it’s time to generate your first image. While Scenario offers standard text-to-image endpoints, high-performance third-party models (such as **Flux** or **Imagen**) utilize a **Custom Generation** workflow.

Unlike standard generation, this process is **asynchronous**. You will submit a request to start a “job,” receive a Job ID, and then poll the API to retrieve your final image asset.

## 1. Initiate the Generation

To generate an image, you will send a `POST` request to the custom generation endpoint. You must specify a `modelId` (e.g., `model_flux-1-dev` or `model_imagen4-ultra`) in the URL.

### The Request

- **Endpoint:** `POST https://api.cloud.scenario.com/v1/generate/custom/{modelId}`
- **Key Parameters:**
- `prompt`: A detailed description of the image you want.
- `aspectRatio`: The desired dimensions (e.g., `"1:1"`, `"16:9"`, `"9:16"`).

**Example cURL:**

Terminal window

```
curl -X POST "https://api.cloud.scenario.com/v1/generate/custom/model_flux-1-dev" \
     -u "YOUR_API_KEY:YOUR_API_SECRET" \
     -H "Content-Type: application/json" \
     -d '{
       "prompt": "A cinematic shot of a futuristic neon city in the rain, hyper-realistic, 8k",
       "aspectRatio": "16:9"
     }'
```

**The Response:** The API will return a `job` object containing a `jobId`.

```
{
  "job": {
    "id": "job_abcd1234efgh",
    "status": "queued"
  }
}
```

---

## 2. Poll for the Result

Because the image is being processed in the cloud, you need to check the status of your `jobId` until it is completed.

**Endpoint:** `GET https://api.cloud.scenario.com/v1/jobs/{jobId}`

**Example Request:**

Terminal window

```
curl -u "YOUR_API_KEY:YOUR_API_SECRET" \
     "https://api.cloud.scenario.com/v1/jobs/job_abcd1234efgh"
```

Keep polling every 2–3 seconds until the `status` changes from `"queued"` or `"in-progress"` to `"success"`.

---

## 3. Retrieve Your Image

Once the job status is `"success"`, the response will include an `images` array (or `result` object) containing the URL of your generated asset.

**Success Response Snippet:**

```
{
  "job": {
    "id": "job_abcd1234efgh",
    "status": "success",
    "result": {
      "images": [
        {
          "url": "https://media.scenario.com/assets/asset_xyz123.png"
        }
      ]
    }
  }
}
```

## What’s Next?

Now that you’ve successfully generated an image using a third-party model, you can:

- **Explore advanced parameters:** Check the specific [Parameters Reference](/get-started/generation/3d-model-generation/index.md) for the model you are using.
- **List available models:** Use `GET /v1/models` to see which other models you can use.
- **Train your own:** Move to Step 4 to learn how to train a model on your own specific art style.