> ## Documentation Index
> Fetch the complete documentation index at: https://docs.modelslab.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Text to Image
> Create high quality images from text using the FLUX model.
Generate photorealistic images from text descriptions using the FLUX model. FLUX excels at creating highly detailed, realistic images with excellent prompt adherence.
## Request
Make a `POST` request to the endpoint below with the required parameters.
```bash theme={null}
POST https://modelslab.com/api/v6/images/text2img
```
## Parameters
Your API key for authentication. Get one from your [dashboard](https://modelslab.com/dashboard/api-keys).
The model to use for generation. Use `flux` for the FLUX model.
Text description of the image you want to generate. Be specific and detailed for best results.
Things you don't want in the image. Example: "blurry, low quality, distorted"
Image width in pixels. Must be between 256-1024 and divisible by 8.
Image height in pixels. Must be between 256-1024 and divisible by 8.
Number of images to generate. Range: 1-4.
Number of denoising steps. Higher values produce more detailed images but take longer. Range: 20-50.
How closely to follow the prompt. Higher values = more literal interpretation. Range: 1-20.
Enable NSFW content filter. Values: "yes" or "no".
Random seed for reproducible results. Use the same seed with same parameters to get identical images.
Number of CLIP layers to skip. Affects style interpretation. Range: 1-4.
URL to receive a POST request when generation completes. See [Webhooks](/webhooks).
Your custom identifier to track this request. Returned in response and webhook.
## Request Example
```json Body theme={null}
{
"key": "your_api_key",
"model_id": "flux",
"prompt": "ultra realistic close up portrait ((beautiful pale cyberpunk female with heavy black eyeliner)), blue eyes, shaved side haircut, hyper detail, cinematic lighting, magic neon, dark red city, Canon EOS R3, nikon, f/1.4, ISO 200, 1/160s, 8K, RAW, unedited, symmetrical balance, in-frame, 8K",
"negative_prompt": "blurry, low quality, distorted, deformed",
"width": 512,
"height": 512,
"samples": 1,
"num_inference_steps": 31,
"safety_checker": "no",
"seed": null,
"guidance_scale": 7.5,
"clip_skip": 2,
"webhook": null,
"track_id": null
}
```
```python Python theme={null}
import requests
response = requests.post(
"https://modelslab.com/api/v6/images/text2img",
json={
"key": "your_api_key",
"model_id": "flux",
"prompt": "ultra realistic portrait of a cyberpunk woman, blue eyes, cinematic lighting, 8K",
"negative_prompt": "blurry, low quality, distorted",
"width": 512,
"height": 512,
"samples": 1,
"num_inference_steps": 31,
"guidance_scale": 7.5
}
)
data = response.json()
print(data)
```
```javascript JavaScript theme={null}
const response = await fetch("https://modelslab.com/api/v6/images/text2img", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
key: "your_api_key",
model_id: "flux",
prompt: "ultra realistic portrait of a cyberpunk woman, blue eyes, cinematic lighting, 8K",
negative_prompt: "blurry, low quality, distorted",
width: 512,
height: 512,
samples: 1,
num_inference_steps: 31,
guidance_scale: 7.5
})
});
const data = await response.json();
console.log(data);
```
```bash cURL theme={null}
curl -X POST "https://modelslab.com/api/v6/images/text2img" \
-H "Content-Type: application/json" \
-d '{
"key": "your_api_key",
"model_id": "flux",
"prompt": "ultra realistic portrait of a cyberpunk woman, blue eyes, cinematic lighting, 8K",
"negative_prompt": "blurry, low quality, distorted",
"width": 512,
"height": 512,
"samples": 1,
"num_inference_steps": 31,
"guidance_scale": 7.5
}'
```
## Response
### Success Response
When the image is generated successfully:
```json theme={null}
{
"status": "success",
"generationTime": 2.45,
"id": 12345678,
"output": [
"https://pub-3626123a908346a7a8be8d9295f44e26.r2.dev/generations/abc123.png"
],
"proxy_links": {
"0": "https://cdn.modelslab.com/generations/abc123.png"
},
"meta": {
"prompt": "ultra realistic portrait...",
"model_id": "flux",
"width": 512,
"height": 512,
"seed": 12345
}
}
```
### Processing Response
For complex requests, you may receive a processing status:
```json theme={null}
{
"status": "processing",
"id": 12345678,
"eta": 10,
"message": "Your request is being processed"
}
```
When you receive a processing response, use the [Fetch endpoint](/image-generation/community-models/dreambooth-fetch-queue-image) to poll for results.
### Error Response
```json theme={null}
{
"status": "error",
"message": "prompt is required"
}
```
## Response Fields
| Field | Type | Description |
| ---------------- | ------- | -------------------------------------- |
| `status` | string | "success", "processing", or "error" |
| `id` | integer | Unique request identifier |
| `output` | array | URLs of generated images |
| `proxy_links` | object | CDN-optimized URLs for faster delivery |
| `generationTime` | float | Time taken to generate (seconds) |
| `meta` | object | Generation parameters used |
## Tips for Better Results
FLUX responds well to detailed, specific prompts. Include:
* Subject description
* Style (photorealistic, cinematic, etc.)
* Lighting details
* Camera settings (for photorealistic images)
Always include a negative prompt to avoid common issues:
```
"blurry, low quality, distorted, deformed, ugly, bad anatomy"
```
* **Lower (3-7)**: More creative, varied results
* **Higher (8-15)**: More literal prompt adherence
* **Default (7.5)**: Good balance for most use cases
## Related Endpoints
Transform existing images with FLUX
Retrieve results for processing requests
## OpenAPI
````yaml POST /text2img
openapi: 3.1.0
info:
title: ModelsLab Dreambooth Image Generation API
description: >-
A focused API for AI-driven image generation using Dreambooth models,
including text-to-image, image-to-image, inpainting, and model management
capabilities.
license:
name: MIT
version: 6.0.0
servers:
- url: https://modelslab.com/api/v6/images
description: Dreambooth Image Generation API v6 server
security: []
paths:
/text2img:
post:
summary: Text-to-Image
description: >-
Creates an image from a text prompt using Dreambooth models. This
endpoint supports both single and multiple LoRA models for enhanced
results.
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/TextToImageRequest'
responses:
'200':
description: Text-to-image generation response
content:
application/json:
schema:
oneOf:
- $ref: '#/components/schemas/ImageGenerationSuccessResponse'
- $ref: '#/components/schemas/ImageGenerationProcessingResponse'
- $ref: '#/components/schemas/ImageGenerationErrorResponse'
'400':
$ref: '#/components/responses/StandardErrorResponse'
'401':
$ref: '#/components/responses/StandardErrorResponse'
'500':
$ref: '#/components/responses/StandardErrorResponse'
components:
schemas:
TextToImageRequest:
type: object
allOf:
- $ref: '#/components/schemas/CommonImageGenerationParams'
required:
- key
- model_id
- prompt
properties:
model_id:
type: string
description: >-
The ID of the model to be used. It can be a public model or one you
have trained.
multi_lingual:
type: boolean
default: false
description: >-
Allow multilingual prompts. Set to 'true' if using a language other
than English.
upscale:
type: boolean
default: false
description: Set to 'true' to upscale the image resolution two times (2x).
highres_fix:
type: boolean
default: false
description: Enable high-resolution fix for generated images.
ImageGenerationSuccessResponse:
type: object
properties:
status:
type: string
enum:
- success
description: Status of the image generation.
generationTime:
type: number
description: Time taken to generate the image in seconds.
id:
type: integer
description: Unique identifier for the image generation request.
output:
type: array
items:
type: string
format: uri
description: Array of generated image URLs.
proxy_links:
type: array
items:
type: string
format: uri
description: Array of proxy image URLs.
meta:
type: object
description: Metadata about the image generation including all parameters used.
additionalProperties: true
nsfw_content_detected:
type: boolean
description: Indicates if NSFW content was detected in the generated image.
webhook_status:
type: string
description: Status of the webhook notification.
tip:
type: string
description: Additional information or tips for the user.
ImageGenerationProcessingResponse:
type: object
properties:
status:
type: string
enum:
- processing
description: Status of the image generation.
tip:
type: string
description: Information about faster processing options.
tip_1:
type: string
description: Additional tip, usually pointing to the fetch API.
eta:
type: number
description: Estimated time for completion in seconds.
message:
type: string
description: Processing status message.
webhook_status:
type: string
description: Status of the webhook notification.
fetch_result:
type: string
format: uri
description: URL to fetch the result when processing.
id:
type: integer
description: Unique identifier for the image generation request.
output:
type: array
items:
type: string
description: Empty array during processing.
meta:
type: object
description: Metadata about the image generation including all parameters used.
additionalProperties: true
future_links:
type: array
items:
type: string
format: uri
description: Array of future image URLs for queued requests.
ImageGenerationErrorResponse:
type: object
required:
- status
- message
properties:
status:
type: string
enum:
- error
message:
type: string
description: Error message description.
CommonImageGenerationParams:
type: object
properties:
key:
type: string
description: Your API Key used for request authorization.
prompt:
type: string
description: A text description of what you want in the generated image.
negative_prompt:
type: string
description: Items you don't want in the image.
enhance_prompt:
type: boolean
default: 'yes'
description: >-
If true, the prompt will be enhanced for better results. Default is
(true).
width:
type: integer
default: 512
maximum: 1024
description: The width of the image in pixels.
height:
type: integer
default: 512
maximum: 1024
description: The height of the image in pixels.
samples:
type: integer
default: 1
maximum: 4
description: The number of images to be returned in response. Maximum is 4.
num_inference_steps:
type: integer
default: 31
minimum: 1
maximum: 20
description: >-
The number of denoising steps. Values range from 1 to 20, and any
value above 20 will be capped at 20.
safety_checker:
type: boolean
default: false
description: >-
A checker for NSFW images. If detected, such images will be replaced
by a blank image.
safety_checker_type:
$ref: '#/components/schemas/SafetyCheckerTypeEnum'
seed:
oneOf:
- type: integer
- type: 'null'
description: >-
Seed for reproducible results. The same seed generates the same
image. Pass null for a random number.
guidance_scale:
type: number
default: 7.5
minimum: 1
maximum: 20
description: Scale for classifier-free guidance.
use_karras_sigmas:
type: boolean
default: true
description: Use Karras sigmas to generate images. Produces nice results.
algorithm_type:
type: string
enum:
- none
- dpmsolver+++
default: none
description: Used in DPMSolverMultistepScheduler scheduler.
vae:
type: string
nullable: true
description: Use a custom VAE for generating images. Default is null.
lora_strength:
type: integer
description: >-
Strength of the LoRa model(s). If multiple LoRa models, provide
comma-separated values (0.1 to 1).
lora_model:
type: string
description: >-
LoRa model ID(s). Multiple LoRa models are supported; pass
comma-separated values (e.g., 'contrast-fix,yae-miko-genshin').
clip_skip:
type: integer
default: 2
minimum: 1
maximum: 8
description: Number of CLIP layers to skip.
base64:
type: boolean
default: false
description: >-
If true, response output is base64 string. Input images can also be
base64.
temp:
type: boolean
default: false
description: If true, stores image in temporary storage (cleaned every 24 hours).
webhook:
type: string
format: uri
description: URL to receive a POST API call once image generation is complete.
track_id:
type: string
description: Unique ID used in webhook response to identify the request.
ip_adapter_id:
$ref: '#/components/schemas/IPAdapterIDEnum'
ip_adapter_scale:
type: number
minimum: 0
maximum: 1
description: Scale for the IP adapter (0 to 1).
ip_adapter_image:
type: string
format: uri
description: Valid image URL for IP adapter.
scheduler:
$ref: '#/components/schemas/SchedulerEnum'
SafetyCheckerTypeEnum:
type: string
enum:
- blur
- sensitive_content_text
- pixelate
- black
default: sensitive_content_text
description: How to modify the image if NSFW content is found.
IPAdapterIDEnum:
type: string
enum:
- ip-adapter_sdxl
- ip-adapter_sd15
- ip-adapter-plus-face_sd15
- ip-adapter-plus_sdxl_vit-h
- ip-adapter-plus-face_sdxl_vit-h
description: IP adapter ID.
SchedulerEnum:
type: string
enum:
- DDPMScheduler
- DDIMScheduler
- PNDMScheduler
- LMSDiscreteScheduler
- EulerDiscreteScheduler
- EulerAncestralDiscreteScheduler
- DPMSolverMultistepScheduler
- HeunDiscreteScheduler
- KDPM2DiscreteScheduler
- DPMSolverSinglestepScheduler
- KDPM2AncestralDiscreteScheduler
- UniPCMultistepScheduler
- DDIMInverseScheduler
- DEISMultistepScheduler
- IPNDMScheduler
- KarrasVeScheduler
- ScoreSdeVeScheduler
- LCMScheduler
description: Available schedulers for image generation.
responses:
StandardErrorResponse:
description: >-
Bad request (invalid parameters), unauthorized (invalid API key), or
internal server error.
content:
application/json:
schema:
$ref: '#/components/schemas/ImageGenerationErrorResponse'
````