Skip to main content

Text to Image Endpoint

Overview​

The Text to Image API is used to create image from text based using flux model. This endpoint generates and returns an image from a text passed in the request body.

Open in Playground 🚀

Text to image endpoint result

Request​

--request POST 'https://modelslab.com/api/v6/images/text2img' \

Make a POST request to https://modelslab.com/api/v6/images/text2img endpoint and pass the required parameters as a request body.

Watch the how-to video to see it in action.

Body Attributes​

ParameterDescriptionType
keyYour API Key used for request authorization.key
model_idThe ID of the model. Obtain the model ID from the provided source.ID
promptA text description of the things you want in the generated image.string
negative_promptItems you don't want in the image.string
widthThe width of the image. Maximum value is 1024.integer
heightThe height of the image. Maximum value is 1024.integer
samplesThe number of images to be returned in response. The maximum value is 4.integer
num_inference_stepsThe number of denoising steps. values range from 1 to 20, and any value above 20 will be capped at 20.
safety_checkerA checker for NSFW images. If detected, such images will be replaced by a blank image. Default is "yes". Options: "yes"/"no".string
safety_checker_typeModify image if NSFW images are found. Default is "sensitive_content_text". Options: "blur", "sensitive_content_text", "pixelate", "black".string
seedUsed to reproduce results. The same seed will generate the same image again. Pass null for a random number.integer
guidance_scaleScale for classifier-free guidance. Minimum value is 1, maximum value is 20.integer
injected_noise_alphaControls the amount of noise injected into the image generation process. This parameter affects the balance between maintaining the original structure and introducing randomness in the generated images. Default Value: 0.75. Range from 0.0 to 1.0float
sigma_thresholdDetermines the noise level threshold at which the image generation process changes its behavior. This parameter affects how the model handles different noise levels during sampling. Default Value: 0.45. Range from 0.0 to 1.0float
true_cfg_scaleModifies the effective classifier-free guidance scale. This parameter allows for fine-tuning of how strongly the model adheres to the provided prompts. Default Value: 1.0. Valid Range: 0.0 to 10.0float
use_cfg_zero_starEnables or disables the use of a zero CFG variant for improved stability. This alternative approach can help reduce certain artifacts in the generation process. Default Value: "yes", Valid Values: "yes" or "noyes/no
use_zero_initControls whether to initialize the diffusion process with zeros instead of random noise. This can affect the starting point of the generation process. Default Value: "no". Valid Values: "yes" or "noyes/no
zero_stepsSpecifies the number of initial steps to use zero noise initialization if use_zero_init is enabled. Determines how long the model uses zero initialization before switching to standard noise handling. Default Value: 0. Valid Range: 0 to maximum sampling steps (typically 0-50)integer
base64Get response as a base64 string. Default is no. Options: yes/no.string
webhookSet a URL to receive a POST API call once image generation is completeURL
track_idThis ID is returned in the response to the webhook API call and is used to identify the webhook request.ID
tempCreate a temporary image link valid for 24 hours. Options: "yes"/"no".string
Open in Playground 🚀

Flux generations​

Flux Model (Full HD Quality)​

Use Case: High-quality Full HD image generation
Generation Time: 4-5 seconds
Description: The Flux model delivers Full HD images with enhanced visual quality, offering a balance between speed and resolution for high-quality outputs.

Body​

Body
{
  "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",
  "width": "512",
  "height": "512",
  "samples": "1",
  "num_inference_steps": "31",
  "safety_checker": "no",
  "enhance_prompt": "yes",
  "seed": null,
  "guidance_scale": 7.5,
  "tomesd": "yes",
  "clip_skip": "2",
  "vae": null,
  "webhook": null,
  "track_id": null
}

Request​

var myHeaders = new Headers();
myHeaders.append("Content-Type", "application/json");

var raw = JSON.stringify({
  "key": "",
  "model_id": "flux",
  "prompt": "actual 8K portrait photo of gareth person, portrait, happy colors, bright eyes, clear eyes, warm smile, smooth soft skin, big dreamy eyes, beautiful intricate colored hair, symmetrical, anime wide eyes, soft lighting, detailed face, by makoto shinkai, stanley artgerm lau, wlop, rossdraws, concept art, digital painting, looking into camera",
  "width": "512",
  "height": "512",
  "samples": "1",
  "num_inference_steps": "30",
  "safety_checker": "no",
  "enhance_prompt": "yes",
  "seed": null,
  "guidance_scale": 7.5,
  "self_attention": "no",
  "vae": null,
  "webhook": null,
  "track_id": null
});

var requestOptions = {
  method: 'POST',
  headers: myHeaders,
  body: raw,
  redirect: 'follow'
};

fetch("https://modelslab.com/api/v6/images/text2img", requestOptions)
  .then(response => response.text())
  .then(result => console.log(result))
  .catch(error => console.log('error', error));

Response​

{
    "status": "success",
    "generationTime": 1.9110066890716553,
    "id": 12384052,
    "output": [
      "https://pub-8b49af329fae499aa563997f5d4068a4.r2.dev/generations/4118bb93-2c49-4d1d-9837-f43a0801e42e-0.png"
    ],
    "meta": {
       "base64": "no",
        "clip_skip": null,
        "enhance_prompt": "yes",
        "file_prefix": "29396946-e6e7-429d-9eda-7b6095868653",
        "guidance_scale": 7.5,
        "height": 1024,
        "id": "134693666",
        "injected_noise_alpha": 0.85,
        "instant_response": "no",
        "model_id": "flux",
        "n_samples": 1,
        "negative_prompt": "ugly, low contrast, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, signature, cut off, low contrast, underexposed, overexposed, bad art, beginner, amateur, distorted face, blurry, draft, grainy",
        "num_inference_steps": 12,
        "opacity": 0.8,
        "padding_down": 10,
        "padding_right": 10,
        "prompt": "ultra realistic close up portrait ((beautiful pale cyberpunk female with heavy black eyeliner))",
        "rescale": "yes",
        "safety_checker": "no",
        "safety_checker_type": "black",
        "scale_down": 6,
        "seed": 2611448049,
        "sigma_threshold": 0.45,
        "temp": "no",
        "track_id": null,
        "true_cfg_scale": 1,
        "use_cfg_zero_star": true,
        "use_zero_init": false,
        "watermark": "no",
        "webhook": null,
        "width": 1024,
        "zero_steps": 0
    }
}