Usage with API

[AldeRoberge]

If anyone is having trouble to get this to work with the API, here is the json that worked for me :

{
                                    "prompt": "<PROMPT>", // Your prompt here
                                    "steps": 20,
                                    "model": "dreamshaper_8", // Your model here
                                    "script_name": "ABG Remover",
                                    "script_args": [
                                      true,  
                                      false, 
                                      "",    
                                      "FFFFFF", 
                                      false   
                                    ]
                                  }

Complete example (C#) :

using System.Text;
using System.Text.Json;

namespace ADG_ImageGenerator;

public interface IImageGenerator
{
    public Task<string?> GenerateImage(string prompt);
}

public class ImageGenerator : IImageGenerator
{
    private const string ApiUrl = "http://127.0.0.1:7860";
    private const string Text2ImageEndpoint = "/sdapi/v1/txt2img";

    /// <summary>
    /// Generates an image from the given prompt using the API.
    /// </summary>
    /// <param name="prompt">The prompt to generate the image</param>
    /// <returns>The base64 image string</returns>
    public async Task<string?> GenerateImage(string prompt)
    {
        using (var client = new HttpClient())
        {
            try
            {
                var requestJson = """
                                  {
                                    "prompt": "<PROMPT>",
                                    "steps": 20,
                                    "model": "dreamshaper_8",
                                    "script_name": "ABG Remover",
                                    "script_args": [
                                      true,  
                                      false, 
                                      "",    
                                      "FFFFFF", 
                                      false   
                                    ]
                                  }
                                  """;

                requestJson = requestJson.Replace("<PROMPT>", prompt);

                var jsonContent = new StringContent(requestJson, Encoding.UTF8,
                    "application/json");

                var response = await client.PostAsync($"{ApiUrl}{Text2ImageEndpoint}", jsonContent);

                if (response.IsSuccessStatusCode)
                {
                    var responseBody = await response.Content.ReadAsStringAsync();

                    // Assuming response contains base64 image string under "image" key
                    var jsonResponse = JsonDocument.Parse(responseBody);
                    if (jsonResponse.RootElement.TryGetProperty("images", out var imageProperty))
                    {
                        // Extract the first image from the array of images
                        var base64Image = imageProperty[0].GetString();
                        return base64Image;
                    }

                    Console.WriteLine("Image not found in the response.");
                }
                else
                {
                    Console.WriteLine($"API call failed with status code: {response.StatusCode}");
                }
            }
            catch (Exception ex)
            {
                Console.WriteLine($"Exception occurred: {ex.Message}");
            }
        }

        return null;
    }
}

Maybe we could include the json somewhere in the documentation?

Thank you

[GeekyGhost]

Thanks, I'll take a look at this.