revert changes to UI

crates/api_clients/fal_client/src/requests/webhook/image/enqueue_gpt_image_1_text_to_image_webhook.rs

@@ -0,0 +1,123 @@
+use crate::creds::fal_api_key::FalApiKey;
+use crate::creds::open_ai_api_key::OpenAiApiKey;
+use crate::error::classify_fal_error::classify_fal_error;
+use crate::error::fal_error_plus::FalErrorPlus;
+use fal::prelude::fal_ai::gpt_image_1::edit_image::byok::{gpt_text_to_image, GptTextToImageRequest};
+use fal::webhook::WebhookResponse;
+use reqwest::IntoUrl;
+
+pub struct GptTextToImageByokArgs<'a, V: IntoUrl> {
+  // Request
+  pub prompt: &'a str,
+  pub image_size: GptTextToImageSize,
+  pub num_images: GptTextToImageNumImages,
+  pub quality: GptTextToImageQuality,
+  
+  // Fulfillment
+  pub api_key: &'a FalApiKey,
+  pub openai_api_key: &'a OpenAiApiKey,
+  pub webhook_url: V,
+}
+
+#[derive(Copy, Clone, Debug)]
+pub enum GptTextToImageSize{
+  Auto,
+  Square,
+  Horizontal,
+  Vertical,
+}
+
+#[derive(Copy, Clone, Debug)]
+pub enum GptTextToImageQuality {
+  Auto,
+  Low,
+  Medium,
+  High,
+}
+
+#[derive(Copy, Clone, Debug)]
+pub enum GptTextToImageNumImages{
+  One,
+  Two,
+  Three,
+  Four,
+}
+
+pub async fn enqueue_gpt_image_1_text_to_image_webhook<V: IntoUrl>(
+  args: GptTextToImageByokArgs<'_, V>
+) -> Result<WebhookResponse, FalErrorPlus> {
+
+  // auto, 1024x1024, 1536x1024, 1024x1536
+  let image_size = match args.image_size {
+    GptTextToImageSize::Auto => "auto",
+    GptTextToImageSize::Square => "1024x1024",
+    GptTextToImageSize::Horizontal => "1536x1024",
+    GptTextToImageSize::Vertical => "1024x1536",
+  };
+  
+  let quality = match args.quality {
+    GptTextToImageQuality::Auto => "auto",
+    GptTextToImageQuality::Low => "low",
+    GptTextToImageQuality::Medium => "medium",
+    GptTextToImageQuality::High => "high",
+  };
+
+  let num_images = match args.num_images {
+    GptTextToImageNumImages::One => 1,
+    GptTextToImageNumImages::Two => 2,
+    GptTextToImageNumImages::Three => 3,
+    GptTextToImageNumImages::Four => 4,
+  };
+
+  let request = GptTextToImageRequest {
+    prompt: args.prompt.to_string(),
+    image_size: image_size.to_string(),
+    num_images,
+    quality: quality.to_string(),
+    openai_api_key: args.openai_api_key.0.to_string(),
+  };
+
+  let result = gpt_text_to_image(request)
+      .with_api_key(&args.api_key.0)
+      .queue_webhook(args.webhook_url)
+      .await;
+
+  result.map_err(|err| classify_fal_error(err))
+}
+
+
+#[cfg(test)]
+mod tests {
+  use crate::creds::fal_api_key::FalApiKey;
+  use crate::creds::open_ai_api_key::OpenAiApiKey;
+  use crate::requests::webhook::image::enqueue_gpt_image_1_text_to_image_webhook::{enqueue_gpt_image_1_text_to_image_webhook, GptTextToImageByokArgs, GptTextToImageNumImages, GptTextToImageQuality, GptTextToImageSize};
+  use errors::AnyhowResult;
+  use std::fs::read_to_string;
+
+  #[tokio::test]
+  #[ignore]
+  async fn test_gpt_image_1_edit_image() -> AnyhowResult<()> {
+    // XXX: Don't commit secrets!
+    let secret = read_to_string("/Users/bt/Artcraft/credentials/fal_api_key.txt")?;
+
+    let fal_api_key = FalApiKey::from_str(&secret);
+
+    let secret = read_to_string("/Users/bt/Artcraft/credentials/openai_api_key.txt")?;
+
+    let open_ai_api_key = OpenAiApiKey::from_str(&secret);
+
+    let args = GptTextToImageByokArgs {
+      prompt: "put the man and the ghost on the grassy hill. the man is scared of the friendly ghost.",
+      api_key: &fal_api_key,
+      openai_api_key: &open_ai_api_key,
+      webhook_url: "https://example.com/webhook",
+      image_size: GptTextToImageSize::Horizontal,
+      num_images: GptTextToImageNumImages::One,
+      quality: GptTextToImageQuality::High,
+    };
+
+    let result = enqueue_gpt_image_1_text_to_image_webhook(args).await?;
+
+    Ok(())
+  }
+}

crates/api_clients/fal_client/src/requests/webhook/image/mod.rs

@@ -3,4 +3,5 @@ pub mod enqueue_flux_1_schnell_text_to_image_webhook;
 pub mod enqueue_flux_pro_11_text_to_image_webhook;
 pub mod enqueue_flux_pro_11_ultra_text_to_image_webhook;
 pub mod enqueue_gpt_image_1_edit_image_webhook;
+pub mod enqueue_gpt_image_1_text_to_image_webhook;
 pub mod remove_background_rembg_webhook;

crates/api_clients/storyteller_client/src/generate/image/generate_gpt_image_1_text_to_image.rs

@@ -0,0 +1,19 @@
+use crate::credentials::storyteller_credential_set::StorytellerCredentialSet;
+use crate::error::storyteller_error::StorytellerError;
+use crate::utils::api_host::ApiHost;
+use crate::utils::basic_json_post_request::basic_json_post_request;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::{GenerateGptImage1TextToImageRequest, GenerateGptImage1TextToImageResponse, GPT_IMAGE_1_TEXT_TO_IMAGE_PATH};
+
+
+pub async fn generate_gpt_image_1_text_to_image(
+  api_host: &ApiHost,
+  maybe_creds: Option<&StorytellerCredentialSet>,
+  request: GenerateGptImage1TextToImageRequest,
+) -> Result<GenerateGptImage1TextToImageResponse, StorytellerError> {
+  Ok(basic_json_post_request(
+    api_host,
+    GPT_IMAGE_1_TEXT_TO_IMAGE_PATH,
+    maybe_creds,
+    request,
+  ).await?)
+}

crates/api_clients/storyteller_client/src/generate/image/mod.rs

@@ -3,4 +3,5 @@ pub mod generate_flux_1_dev_text_to_image;
 pub mod generate_flux_1_schnell_text_to_image;
 pub mod generate_flux_pro_11_text_to_image;
 pub mod generate_flux_pro_11_ultra_text_to_image;
+pub mod generate_gpt_image_1_text_to_image;
 pub mod remove_image_background;

crates/desktop/artcraft/src/core/commands/enqueue/image/enqueue_text_to_image_command.rs

@@ -1,6 +1,7 @@
 use crate::core::commands::enqueue::image::generic::handle_image_artcraft::handle_image_artcraft;
 use crate::core::commands::enqueue::image::generic::handle_image_fal::handle_image_fal;
-use crate::core::commands::enqueue::image::gpt_image_1::handle_image_sora::handle_image_sora;
+use crate::core::commands::enqueue::image::gpt_image_1::handle_gpt_image_1::handle_gpt_image_1;
+use crate::core::commands::enqueue::image::gpt_image_1::handle_gpt_image_1_sora::handle_gpt_image_1_sora;
 use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::commands::response::failure_response_wrapper::{CommandErrorResponseWrapper, CommandErrorStatus};
@@ -39,6 +40,9 @@ pub struct EnqueueTextToImageRequest {
 
   /// Aspect ratio.
   pub aspect_ratio: Option<TextToImageSize>,
+
+  /// The number of images to generate.
+  pub number_images: Option<u32>,
 }
 
 // TODO(bt,2025-07-14): Support other aspect ratios / resolutions -
@@ -224,7 +228,16 @@ pub async fn dispatch_request(
       return Err(InternalImageError::NoModelSpecified);
     }
     Some(ImageModel::GptImage1) => {
-      return handle_image_sora(&app, request, sora_creds_manager, sora_task_queue).await;
+      return handle_gpt_image_1(
+        request,
+        app,
+        app_data_root,
+        app_env_configs,
+        provider_priority_store,
+        storyteller_creds_manager,
+        sora_creds_manager, 
+        sora_task_queue,
+      ).await;
     }
     _ => {
       // Fall-through

crates/desktop/artcraft/src/core/commands/enqueue/image/generic/handle_image_artcraft.rs

@@ -9,10 +9,10 @@ use crate::core::state::app_env_configs::app_env_configs::AppEnvConfigs;
 use crate::core::state::data_dir::app_data_root::AppDataRoot;
 use crate::services::storyteller::state::storyteller_credential_manager::StorytellerCredentialManager;
 use anyhow::anyhow;
-use artcraft_api_defs::generate::image::generate_flux_1_dev_text_to_image::{GenerateFlux1DevTextToImageAspectRatio, GenerateFlux1DevTextToImageRequest};
-use artcraft_api_defs::generate::image::generate_flux_1_schnell_text_to_image::{GenerateFlux1SchnellTextToImageAspectRatio, GenerateFlux1SchnellTextToImageRequest};
-use artcraft_api_defs::generate::image::generate_flux_pro_11_text_to_image::{GenerateFluxPro11TextToImageAspectRatio, GenerateFluxPro11TextToImageRequest};
-use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image::{GenerateFluxPro11UltraTextToImageAspectRatio, GenerateFluxPro11UltraTextToImageRequest};
+use artcraft_api_defs::generate::image::generate_flux_1_dev_text_to_image::{GenerateFlux1DevTextToImageAspectRatio, GenerateFlux1DevTextToImageNumImages, GenerateFlux1DevTextToImageRequest};
+use artcraft_api_defs::generate::image::generate_flux_1_schnell_text_to_image::{GenerateFlux1SchnellTextToImageAspectRatio, GenerateFlux1SchnellTextToImageNumImages, GenerateFlux1SchnellTextToImageRequest};
+use artcraft_api_defs::generate::image::generate_flux_pro_11_text_to_image::{GenerateFluxPro11TextToImageAspectRatio, GenerateFluxPro11TextToImageNumImages, GenerateFluxPro11TextToImageRequest};
+use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image::{GenerateFluxPro11UltraTextToImageAspectRatio, GenerateFluxPro11UltraTextToImageNumImages, GenerateFluxPro11UltraTextToImageRequest};
 use enums::common::generation_provider::GenerationProvider;
 use enums::tauri::tasks::task_type::TaskType;
 use fal_client::requests::queue::image_gen::enqueue_flux_pro_11_ultra_text_to_image::{enqueue_flux_pro_11_ultra_text_to_image, FluxPro11UltraTextToImageArgs};
@@ -58,7 +58,7 @@ pub async fn handle_image_artcraft(
       return Err(InternalImageError::NoModelSpecified);
     }
     Some(ImageModel::GptImage1) => {
-      return Err(InternalImageError::AnyhowError(anyhow!("wrong logic: artcraft is handling sora images")));
+      return Err(InternalImageError::AnyhowError(anyhow!("wrong logic: an earlier branch should have handled gpt-image-1 images for Artcraft")));
     }
     Some(ImageModel::Recraft3) => {
       return Err(InternalImageError::AnyhowError(anyhow!("not yet implemented in Artcraft")));
@@ -66,19 +66,25 @@ pub async fn handle_image_artcraft(
     Some(ImageModel::Flux1Dev) => {
       info!("enqueue Flux 1 Dev");
       selected_model = GenerationModel::Flux1Dev;
-      let aspect_ratio = request.aspect_ratio
-          .map(|aspect| match aspect {
-            // TODO(bt,2025-07-14): Support other aspect ratios.
-            TextToImageSize::Tall => GenerateFlux1DevTextToImageAspectRatio::PortraitNineBySixteen,
-            TextToImageSize::Wide => GenerateFlux1DevTextToImageAspectRatio::LandscapeSixteenByNine,
-            TextToImageSize::Auto => GenerateFlux1DevTextToImageAspectRatio::SquareHd,
-            TextToImageSize::Square => GenerateFlux1DevTextToImageAspectRatio::SquareHd,
-          });
       let request = GenerateFlux1DevTextToImageRequest {
         uuid_idempotency_token,
         prompt: request.prompt,
-        aspect_ratio,
-        num_images: None,
+        aspect_ratio: request.aspect_ratio
+            .map(|aspect| match aspect {
+              // TODO(bt,2025-07-14): Support other aspect ratios.
+              TextToImageSize::Tall => GenerateFlux1DevTextToImageAspectRatio::PortraitNineBySixteen,
+              TextToImageSize::Wide => GenerateFlux1DevTextToImageAspectRatio::LandscapeSixteenByNine,
+              TextToImageSize::Auto => GenerateFlux1DevTextToImageAspectRatio::SquareHd,
+              TextToImageSize::Square => GenerateFlux1DevTextToImageAspectRatio::SquareHd,
+            }),
+        num_images: request.number_images
+            .and_then(|num| match num {
+              1 => Some(GenerateFlux1DevTextToImageNumImages::One),
+              2 => Some(GenerateFlux1DevTextToImageNumImages::Two),
+              3 => Some(GenerateFlux1DevTextToImageNumImages::Three),
+              4 => Some(GenerateFlux1DevTextToImageNumImages::Four),
+              _ => None, // NB: use service defaults
+            }),
       };
       let result = generate_flux_1_dev_text_to_image(
         &app_env_configs.storyteller_host,
@@ -99,19 +105,25 @@ pub async fn handle_image_artcraft(
     Some(ImageModel::Flux1Schnell) => {
       info!("enqueue Flux 1 Schnell");
       selected_model = GenerationModel::Flux1Schnell;
-      let aspect_ratio = request.aspect_ratio
-          .map(|aspect| match aspect {
-            // TODO(bt,2025-07-14): Support other aspect ratios.
-            TextToImageSize::Tall => GenerateFlux1SchnellTextToImageAspectRatio::PortraitNineBySixteen,
-            TextToImageSize::Wide => GenerateFlux1SchnellTextToImageAspectRatio::LandscapeSixteenByNine,
-            TextToImageSize::Auto => GenerateFlux1SchnellTextToImageAspectRatio::SquareHd,
-            TextToImageSize::Square => GenerateFlux1SchnellTextToImageAspectRatio::SquareHd,
-          });
       let request = GenerateFlux1SchnellTextToImageRequest {
         uuid_idempotency_token,
         prompt: request.prompt,
-        aspect_ratio,
-        num_images: None,
+        aspect_ratio: request.aspect_ratio
+            .map(|aspect| match aspect {
+              // TODO(bt,2025-07-14): Support other aspect ratios.
+              TextToImageSize::Tall => GenerateFlux1SchnellTextToImageAspectRatio::PortraitNineBySixteen,
+              TextToImageSize::Wide => GenerateFlux1SchnellTextToImageAspectRatio::LandscapeSixteenByNine,
+              TextToImageSize::Square => GenerateFlux1SchnellTextToImageAspectRatio::SquareHd,
+              TextToImageSize::Auto => GenerateFlux1SchnellTextToImageAspectRatio::SquareHd,
+            }),
+        num_images: request.number_images
+            .and_then(|num| match num {
+              1 => Some(GenerateFlux1SchnellTextToImageNumImages::One),
+              2 => Some(GenerateFlux1SchnellTextToImageNumImages::Two),
+              3 => Some(GenerateFlux1SchnellTextToImageNumImages::Three),
+              4 => Some(GenerateFlux1SchnellTextToImageNumImages::Four),
+              _ => None, // NB: use service defaults
+            }),
       };
       let result = generate_flux_1_schnell_text_to_image(
         &app_env_configs.storyteller_host,
@@ -132,19 +144,25 @@ pub async fn handle_image_artcraft(
     Some(ImageModel::FluxPro11) => {
       info!("enqueue Flux Pro 1.1");
       selected_model = GenerationModel::FluxPro11;
-      let aspect_ratio = request.aspect_ratio
-          .map(|aspect| match aspect {
-            // TODO(bt,2025-07-14): Support other aspect ratios.
-            TextToImageSize::Tall => GenerateFluxPro11TextToImageAspectRatio::PortraitNineBySixteen,
-            TextToImageSize::Wide => GenerateFluxPro11TextToImageAspectRatio::LandscapeSixteenByNine,
-            TextToImageSize::Auto => GenerateFluxPro11TextToImageAspectRatio::Square,
-            TextToImageSize::Square => GenerateFluxPro11TextToImageAspectRatio::Square,
-          });
       let request = GenerateFluxPro11TextToImageRequest {
         uuid_idempotency_token,
         prompt: request.prompt,
-        aspect_ratio,
-        num_images: None,
+        aspect_ratio: request.aspect_ratio
+            .map(|aspect| match aspect {
+              // TODO(bt,2025-07-14): Support other aspect ratios.
+              TextToImageSize::Tall => GenerateFluxPro11TextToImageAspectRatio::PortraitNineBySixteen,
+              TextToImageSize::Wide => GenerateFluxPro11TextToImageAspectRatio::LandscapeSixteenByNine,
+              TextToImageSize::Square => GenerateFluxPro11TextToImageAspectRatio::Square,
+              TextToImageSize::Auto => GenerateFluxPro11TextToImageAspectRatio::Square,
+            }),
+        num_images: request.number_images
+            .and_then(|num| match num {
+              1 => Some(GenerateFluxPro11TextToImageNumImages::One),
+              2 => Some(GenerateFluxPro11TextToImageNumImages::Two),
+              3 => Some(GenerateFluxPro11TextToImageNumImages::Three),
+              4 => Some(GenerateFluxPro11TextToImageNumImages::Four),
+              _ => None, // NB: use service defaults
+            }),
       };
       let result = generate_flux_pro_11_text_to_image(
         &app_env_configs.storyteller_host,
@@ -165,19 +183,25 @@ pub async fn handle_image_artcraft(
     Some(ImageModel::FluxPro11Ultra) => {
       info!("enqueue Flux Pro 1.1 Ultra");
       selected_model = GenerationModel::FluxPro11Ultra;
-      let aspect_ratio = request.aspect_ratio
-          .map(|aspect| match aspect {
-            // TODO(bt,2025-07-14): Support other aspect ratios.
-            TextToImageSize::Tall => GenerateFluxPro11UltraTextToImageAspectRatio::PortraitNineBySixteen,
-            TextToImageSize::Wide => GenerateFluxPro11UltraTextToImageAspectRatio::LandscapeSixteenByNine,
-            TextToImageSize::Auto => GenerateFluxPro11UltraTextToImageAspectRatio::Square,
-            TextToImageSize::Square => GenerateFluxPro11UltraTextToImageAspectRatio::Square,
-          });
       let request = GenerateFluxPro11UltraTextToImageRequest {
         uuid_idempotency_token,
         prompt: request.prompt,
-        aspect_ratio,
-        num_images: None,
+        aspect_ratio: request.aspect_ratio
+            .map(|aspect| match aspect {
+              // TODO(bt,2025-07-14): Support other aspect ratios.
+              TextToImageSize::Tall => GenerateFluxPro11UltraTextToImageAspectRatio::PortraitNineBySixteen,
+              TextToImageSize::Wide => GenerateFluxPro11UltraTextToImageAspectRatio::LandscapeSixteenByNine,
+              TextToImageSize::Square => GenerateFluxPro11UltraTextToImageAspectRatio::Square,
+              TextToImageSize::Auto => GenerateFluxPro11UltraTextToImageAspectRatio::Square,
+            }),
+        num_images: request.number_images
+            .and_then(|num| match num {
+              1 => Some(GenerateFluxPro11UltraTextToImageNumImages::One),
+              2 => Some(GenerateFluxPro11UltraTextToImageNumImages::Two),
+              3 => Some(GenerateFluxPro11UltraTextToImageNumImages::Three),
+              4 => Some(GenerateFluxPro11UltraTextToImageNumImages::Four),
+              _ => None, // NB: use service defaults
+            }),
       };
       let result = generate_flux_pro_11_ultra_text_to_image(
         &app_env_configs.storyteller_host,

crates/desktop/artcraft/src/core/commands/enqueue/image/gpt_image_1/handle_gpt_image_1.rs

@@ -0,0 +1,75 @@
+use crate::core::commands::enqueue::image::enqueue_text_to_image_command::{EnqueueTextToImageRequest, TextToImageSize};
+use crate::core::commands::enqueue::image::gpt_image_1::handle_gpt_image_1_artcraft::handle_gpt_image_1_artcraft;
+use crate::core::commands::enqueue::image::gpt_image_1::handle_gpt_image_1_sora::handle_gpt_image_1_sora;
+use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
+use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_artcraft::handle_gpt_image_1_edit_artcraft;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_sora::handle_gpt_image_1_edit_sora;
+use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
+use crate::core::events::basic_sendable_event_trait::BasicSendableEvent;
+use crate::core::events::generation_events::common::{GenerationAction, GenerationModel, GenerationServiceProvider};
+use crate::core::events::generation_events::generation_enqueue_failure_event::GenerationEnqueueFailureEvent;
+use crate::core::model::image_models::ImageModel;
+use crate::core::state::app_env_configs::app_env_configs::AppEnvConfigs;
+use crate::core::state::data_dir::app_data_root::AppDataRoot;
+use crate::core::state::provider_priority::{Provider, ProviderPriorityStore};
+use crate::services::sora::state::sora_credential_manager::SoraCredentialManager;
+use crate::services::sora::state::sora_task_queue::SoraTaskQueue;
+use crate::services::storyteller::state::storyteller_credential_manager::StorytellerCredentialManager;
+use enums::common::generation_provider::GenerationProvider;
+use enums::tauri::tasks::task_type::TaskType;
+use fal_client::requests::queue::image_gen::enqueue_flux_pro_11_ultra_text_to_image::{enqueue_flux_pro_11_ultra_text_to_image, FluxPro11UltraTextToImageArgs};
+use log::{error, info};
+use openai_sora_client::recipes::maybe_upgrade_or_renew_session::maybe_upgrade_or_renew_session;
+use openai_sora_client::recipes::simple_image_gen_with_session_auto_renew::{simple_image_gen_with_session_auto_renew, SimpleImageGenAutoRenewRequest};
+use openai_sora_client::requests::image_gen::common::{ImageSize, NumImages};
+use std::time::Duration;
+use tauri::AppHandle;
+
+pub async fn handle_gpt_image_1(
+  request: EnqueueTextToImageRequest,
+  app: &AppHandle,
+  app_data_root: &AppDataRoot,
+  app_env_configs: &AppEnvConfigs,
+  provider_priority_store: &ProviderPriorityStore,
+  storyteller_creds_manager: &StorytellerCredentialManager,
+  sora_creds_manager: &SoraCredentialManager,
+  sora_task_queue: &SoraTaskQueue,
+) -> Result<TaskEnqueueSuccess, InternalImageError> {
+
+  let priority = provider_priority_store.get_priority()?;
+
+  // TODO: Check if providers are available before proceeding.
+
+  info!("Providers by priority: {:?}", priority);
+
+  for provider in priority.iter() {
+    match provider {
+      Provider::Fal => {
+        // Fallthrough
+        // If in the future we support OpenAI API keys, it's worth considering whether we 
+        // send those requests ourselves, or if we use FAL as an intermediary. FAL makes 
+        // the API nicer to deal with, but the user needs an additional key.
+      }
+      Provider::Artcraft => {
+        info!("Dispatching gpt-image-1 via Artcraft...");
+        return handle_gpt_image_1_artcraft(
+          request,
+          app_env_configs,
+          storyteller_creds_manager
+        ).await;
+      }
+      Provider::Sora => {
+        info!("Dispatching gpt-image-1 via Sora...");
+        return handle_gpt_image_1_sora(
+          request,
+          app,
+          sora_creds_manager,
+          sora_task_queue,
+        ).await;
+      }
+    }
+  }
+
+  Err(InternalImageError::NoProviderAvailable)
+}

crates/desktop/artcraft/src/core/commands/enqueue/image/gpt_image_1/handle_gpt_image_1_artcraft.rs

@@ -0,0 +1,115 @@
+use crate::core::commands::enqueue::image::enqueue_text_to_image_command::{EnqueueTextToImageRequest, TextToImageSize};
+use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
+use crate::core::commands::enqueue::image_edit::enqueue_contextual_edit_image_command::{EditImageQuality, EditImageSize};
+use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
+use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
+use crate::core::events::basic_sendable_event_trait::BasicSendableEvent;
+use crate::core::events::generation_events::common::{GenerationAction, GenerationModel, GenerationServiceProvider};
+use crate::core::events::generation_events::generation_enqueue_failure_event::GenerationEnqueueFailureEvent;
+use crate::core::model::image_models::ImageModel;
+use crate::core::state::app_env_configs::app_env_configs::AppEnvConfigs;
+use crate::core::state::data_dir::app_data_root::AppDataRoot;
+use crate::core::state::provider_priority::ProviderPriorityStore;
+use crate::services::sora::state::sora_credential_manager::SoraCredentialManager;
+use crate::services::sora::state::sora_task_queue::SoraTaskQueue;
+use crate::services::storyteller::state::storyteller_credential_manager::StorytellerCredentialManager;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::{GenerateGptImage1TextToImageImageQuality, GenerateGptImage1TextToImageImageSize, GenerateGptImage1TextToImageNumImages, GenerateGptImage1TextToImageRequest};
+use enums::common::generation_provider::GenerationProvider;
+use enums::tauri::tasks::task_type::TaskType;
+use fal_client::requests::queue::image_gen::enqueue_flux_pro_11_ultra_text_to_image::{enqueue_flux_pro_11_ultra_text_to_image, FluxPro11UltraTextToImageArgs};
+use idempotency::uuid::generate_random_uuid;
+use log::{error, info};
+use openai_sora_client::recipes::maybe_upgrade_or_renew_session::maybe_upgrade_or_renew_session;
+use openai_sora_client::recipes::simple_image_gen_with_session_auto_renew::{simple_image_gen_with_session_auto_renew, SimpleImageGenAutoRenewRequest};
+use openai_sora_client::requests::image_gen::common::{ImageSize, NumImages};
+use std::time::Duration;
+use storyteller_client::generate::image::edit::gpt_image_1_edit_image::gpt_image_1_edit_image;
+use storyteller_client::generate::image::generate_gpt_image_1_text_to_image::generate_gpt_image_1_text_to_image;
+use tauri::AppHandle;
+
+pub async fn handle_gpt_image_1_artcraft(
+  request: EnqueueTextToImageRequest,
+  app_env_configs: &AppEnvConfigs,
+  storyteller_creds_manager: &StorytellerCredentialManager,
+) -> Result<TaskEnqueueSuccess, InternalImageError> {
+
+  let creds = match storyteller_creds_manager.get_credentials()? {
+    Some(creds) => creds,
+    None => {
+      return Err(InternalImageError::NeedsStorytellerCredentials);
+    },
+  };
+
+  info!("Calling Artcraft gpt-image-1 ...");
+
+  let uuid_idempotency_token = generate_random_uuid();
+
+  let image_quality = Some(GenerateGptImage1TextToImageImageQuality::High);
+  
+  //let image_quality = request.image_quality
+  //    .map(|quality| match quality {
+  //      EditImageQuality::Auto => GenerateGptImage1TextToImageImageQuality::Auto,
+  //      EditImageQuality::High => GenerateGptImage1TextToImageImageQuality::High,
+  //      EditImageQuality::Medium => GenerateGptImage1TextToImageImageQuality::Medium,
+  //      EditImageQuality::Low => GenerateGptImage1TextToImageImageQuality::Low,
+  //    });
+
+  let image_size = request.aspect_ratio
+      .map(|size| match size {
+        TextToImageSize::Auto => GenerateGptImage1TextToImageImageSize::Square,
+        TextToImageSize::Square => GenerateGptImage1TextToImageImageSize::Square,
+        TextToImageSize::Tall => GenerateGptImage1TextToImageImageSize::Vertical,
+        TextToImageSize::Wide => GenerateGptImage1TextToImageImageSize::Horizontal,
+      });
+
+  let num_images = match request.number_images {
+    None => None,
+    Some(1) => Some(GenerateGptImage1TextToImageNumImages::One),
+    Some(2) => Some(GenerateGptImage1TextToImageNumImages::Two),
+    Some(3) => Some(GenerateGptImage1TextToImageNumImages::Three),
+    Some(4) => Some(GenerateGptImage1TextToImageNumImages::Four),
+    // TODO: Error
+    //Some(other) => {
+    //  return Err(InternalImageError::InvalidNumberOfRequestedImages {
+    //    min: 1,
+    //    max: 4,
+    //    requested: other,
+    //  });
+    //},
+    _ => Some(GenerateGptImage1TextToImageNumImages::One), // Default to one image if invalid number
+  };
+
+  let request = GenerateGptImage1TextToImageRequest {
+    uuid_idempotency_token,
+    prompt: request.prompt.clone(),
+    image_size,
+    num_images,
+    image_quality,
+  };
+
+  let result = generate_gpt_image_1_text_to_image(
+    &app_env_configs.storyteller_host,
+    Some(&creds),
+    request,
+  ).await;
+
+  let job_id = match result {
+    Ok(enqueued) => {
+      // TODO(bt,2025-07-05): Enqueue job token?
+      info!("Successfully enqueued Artcraft gpt-image-1. Job token: {}", 
+        enqueued.inference_job_token);
+      enqueued.inference_job_token
+    }
+    Err(err) => {
+      error!("Failed to use Artcraft gpt-image-1: {:?}", err);
+      return Err(InternalImageError::StorytellerError(err));
+    }
+  };
+
+  Ok(TaskEnqueueSuccess {
+    provider: GenerationProvider::Artcraft,
+    model: Some(GenerationModel::GptImage1),
+    provider_job_id: Some(job_id.to_string()),
+    task_type: TaskType::ImageGeneration,
+  })
+}

crates/desktop/artcraft/src/core/commands/enqueue/image/gpt_image_1/handle_gpt_image_1_sora.rs

@@ -19,9 +19,9 @@ use tauri::AppHandle;
 
 const SORA_SIMPLE_IMAGE_GEN_TIMEOUT : Duration = Duration::from_millis(1000 * 30); // 30 seconds
 
-pub async fn handle_image_sora(
-  app: &AppHandle,
+pub async fn handle_gpt_image_1_sora(
   request: EnqueueTextToImageRequest,
+  app: &AppHandle,
   sora_creds_manager: &SoraCredentialManager,
   sora_task_queue: &SoraTaskQueue,
 ) -> Result<TaskEnqueueSuccess, InternalImageError> {
@@ -55,10 +55,19 @@ pub async fn handle_image_sora(
     Some(TextToImageSize::Wide) => ImageSize::Wide,
   };
 
+  let num_images = match request.number_images {
+    None => NumImages::One,
+    Some(1) => NumImages::One,
+    Some(2) => NumImages::Two,
+    Some(3) => NumImages::Two, // NB: There is no option for three images!
+    Some(4) => NumImages::Four,
+    _ => NumImages::One,
+  };
+
   let result =
       simple_image_gen_with_session_auto_renew(SimpleImageGenAutoRenewRequest {
         prompt: prompt.to_string(),
-        num_images: NumImages::One,
+        num_images,
         image_size: size,
         credentials: &creds,
         request_timeout: Some(SORA_SIMPLE_IMAGE_GEN_TIMEOUT),

crates/desktop/artcraft/src/core/commands/enqueue/image/gpt_image_1/mod.rs

@@ -1 +1,3 @@
-pub mod handle_image_sora;
+pub (crate) mod handle_gpt_image_1_artcraft;
+pub (crate) mod handle_gpt_image_1_sora;
+pub (super) mod handle_gpt_image_1;

crates/desktop/artcraft/src/core/commands/enqueue/image_bg_removal/enqueue_image_bg_removal_command.rs

@@ -3,7 +3,7 @@ use crate::core::commands::enqueue::image::internal_image_error::InternalImageEr
 use crate::core::commands::enqueue::image_bg_removal::errors::InternalBgRemovalError;
 use crate::core::commands::enqueue::image_bg_removal::generic::handle_generic_bg_removal::handle_generic_bg_removal;
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1::handle_gpt_image_1;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit::handle_gpt_image_1_edit;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::commands::response::failure_response_wrapper::{CommandErrorResponseWrapper, CommandErrorStatus};
 use crate::core::commands::response::shorthand::{Response, ResponseOrErrorType};

crates/desktop/artcraft/src/core/commands/enqueue/image_bg_removal/generic/handle_generic_bg_removal.rs

@@ -4,8 +4,8 @@ use crate::core::commands::enqueue::image_bg_removal::generic::handle_generic_bg
 use crate::core::commands::enqueue::image_bg_removal::generic::handle_generic_bg_removal_fal::handle_generic_bg_removal_fal;
 use crate::core::commands::enqueue::image_edit::enqueue_contextual_edit_image_command::EnqueueContextualEditImageCommand;
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_artcraft::handle_gpt_image_1_artcraft;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_sora::handle_gpt_image_1_sora;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_artcraft::handle_gpt_image_1_edit_artcraft;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_sora::handle_gpt_image_1_edit_sora;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::state::app_env_configs::app_env_configs::AppEnvConfigs;
 use crate::core::state::data_dir::app_data_root::AppDataRoot;

crates/desktop/artcraft/src/core/commands/enqueue/image_edit/enqueue_contextual_edit_image_command.rs

@@ -1,7 +1,7 @@
 use crate::core::commands::enqueue::image::enqueue_text_to_image_command::{EnqueueTextToImageRequest, EnqueueTextToImageSuccessResponse};
 use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1::handle_gpt_image_1;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit::handle_gpt_image_1_edit;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::commands::response::failure_response_wrapper::{CommandErrorResponseWrapper, CommandErrorStatus};
 use crate::core::commands::response::shorthand::{Response, ResponseOrErrorType};
@@ -208,7 +208,7 @@ pub async fn handle_request(
       return Err(InternalContextualEditImageError::NoModelSpecified)
     }
     Some(ContextualImageEditModel::GptImage1) => {
-      handle_gpt_image_1(
+      handle_gpt_image_1_edit(
         request,
         app,
         app_data_root,

crates/desktop/artcraft/src/core/commands/enqueue/image_edit/gpt_image_1/handle_gpt_image_1_edit.rs

@@ -1,7 +1,7 @@
 use crate::core::commands::enqueue::image_edit::enqueue_contextual_edit_image_command::EnqueueContextualEditImageCommand;
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_artcraft::handle_gpt_image_1_artcraft;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_sora::handle_gpt_image_1_sora;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_artcraft::handle_gpt_image_1_edit_artcraft;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit_sora::handle_gpt_image_1_edit_sora;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::state::app_env_configs::app_env_configs::AppEnvConfigs;
 use crate::core::state::data_dir::app_data_root::AppDataRoot;
@@ -16,7 +16,7 @@ use tauri::AppHandle;
 
 pub(super) const MAX_IMAGES: usize = 10;
 
-pub async fn handle_gpt_image_1(
+pub async fn handle_gpt_image_1_edit(
   request: &EnqueueContextualEditImageCommand,
   app: &AppHandle,
   app_data_root: &AppDataRoot,
@@ -44,8 +44,8 @@ pub async fn handle_gpt_image_1(
         // the API nicer to deal with, but the user needs an additional key.
       }
       Provider::Artcraft => {
-        info!("Dispatching gpt-image-1 via Artcraft...");
-        return handle_gpt_image_1_artcraft(
+        info!("Dispatching gpt-image-1 (edit) via Artcraft...");
+        return handle_gpt_image_1_edit_artcraft(
           request, 
           app,
           app_data_root,
@@ -54,8 +54,8 @@ pub async fn handle_gpt_image_1(
         ).await;
       }
       Provider::Sora => {
-        info!("Dispatching gpt-image-1 via Sora...");
-        return handle_gpt_image_1_sora(
+        info!("Dispatching gpt-image-1 (edit) via Sora...");
+        return handle_gpt_image_1_edit_sora(
           request, 
           app,
           app_data_root,

crates/desktop/artcraft/src/core/commands/enqueue/image_edit/gpt_image_1/handle_gpt_image_1_edit_artcraft.rs

@@ -2,7 +2,7 @@ use crate::core::commands::enqueue::image::enqueue_text_to_image_command::Enqueu
 use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
 use crate::core::commands::enqueue::image_edit::enqueue_contextual_edit_image_command::{EditImageQuality, EditImageSize, EnqueueContextualEditImageCommand};
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1::MAX_IMAGES;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit::MAX_IMAGES;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::events::basic_sendable_event_trait::BasicSendableEvent;
 use crate::core::events::generation_events::common::{GenerationAction, GenerationModel, GenerationServiceProvider};
@@ -36,7 +36,7 @@ use tauri::AppHandle;
 use enums::common::generation_provider::GenerationProvider;
 use enums::tauri::tasks::task_type::TaskType;
 
-pub async fn handle_gpt_image_1_artcraft(
+pub async fn handle_gpt_image_1_edit_artcraft(
   request: &EnqueueContextualEditImageCommand,
   app: &AppHandle,
   app_data_root: &AppDataRoot,
@@ -51,7 +51,7 @@ pub async fn handle_gpt_image_1_artcraft(
     },
   };
 
-  info!("Calling Artcraft gpt-image-1 ...");
+  info!("Calling Artcraft gpt-image-1 (edit) ...");
 
   let uuid_idempotency_token = generate_random_uuid();
   

crates/desktop/artcraft/src/core/commands/enqueue/image_edit/gpt_image_1/handle_gpt_image_1_edit_sora.rs

@@ -2,7 +2,7 @@ use crate::core::commands::enqueue::image::enqueue_text_to_image_command::Enqueu
 use crate::core::commands::enqueue::image::internal_image_error::InternalImageError;
 use crate::core::commands::enqueue::image_edit::enqueue_contextual_edit_image_command::{EditImageSize, EnqueueContextualEditImageCommand, EnqueueContextualEditImageErrorType};
 use crate::core::commands::enqueue::image_edit::errors::InternalContextualEditImageError;
-use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1::MAX_IMAGES;
+use crate::core::commands::enqueue::image_edit::gpt_image_1::handle_gpt_image_1_edit::MAX_IMAGES;
 use crate::core::commands::enqueue::task_enqueue_success::TaskEnqueueSuccess;
 use crate::core::commands::response::failure_response_wrapper::{CommandErrorResponseWrapper, CommandErrorStatus};
 use crate::core::events::basic_sendable_event_trait::BasicSendableEvent;
@@ -48,7 +48,7 @@ const SORA_IMAGE_REMIX_TIMEOUT: Duration = Duration::from_millis(1000 * 30); //
 
 const DEFAULT_ASPECT_RATIO : EditImageSize = EditImageSize::Square;
 
-pub async fn handle_gpt_image_1_sora(
+pub async fn handle_gpt_image_1_edit_sora(
   request: &EnqueueContextualEditImageCommand,
   app: &AppHandle,
   app_data_root: &AppDataRoot,

crates/desktop/artcraft/src/core/commands/enqueue/image_edit/gpt_image_1/mod.rs

@@ -1,3 +1,3 @@
-pub mod handle_gpt_image_1;
-pub mod handle_gpt_image_1_artcraft;
-pub mod handle_gpt_image_1_sora;
+pub mod handle_gpt_image_1_edit;
+pub mod handle_gpt_image_1_edit_artcraft;
+pub mod handle_gpt_image_1_edit_sora;

crates/lib/files/images/Cargo.toml

@@ -26,7 +26,7 @@ errors = { workspace = true}
 mimetypes = { workspace = true }
 
 # External
-image = "0.25.6"
+image = { version = "0.25.6" , features = ["jpeg", "png", "tiff", "webp"] }
 
 [dev-dependencies]
 anyhow = { workspace = true }

crates/lib/files/images/src/encoding/mod.rs

@@ -0,0 +1 @@
+pub mod webp_bytes_to_png_bytes;

crates/lib/files/images/src/encoding/webp_bytes_to_png_bytes.rs

@@ -0,0 +1,21 @@
+use crate::error::ImagesError;
+use image::{ImageFormat, ImageReader};
+use std::io::Cursor;
+
+pub fn webp_bytes_to_png_bytes(
+    webp_bytes: &[u8],
+) -> Result<Vec<u8>, ImagesError> {
+
+  let mut reader = ImageReader::new(Cursor::new(webp_bytes));
+  
+  reader.set_format(ImageFormat::WebP); // Default to WebP format (as fallback).
+  
+  let image = reader
+      .with_guessed_format()? // NB: Can raise IoError.
+      .decode()?;
+  
+  let mut output_bytes: Vec<u8> = Vec::new();
+  image.write_to(&mut Cursor::new(&mut output_bytes), ImageFormat::Png)?;
+
+  Ok(output_bytes)
+}

crates/lib/files/images/src/error.rs

@@ -0,0 +1,31 @@
+use std::error::Error;
+use std::fmt::{Display, Formatter};
+
+#[derive(Debug)]
+pub enum ImagesError {
+  IoError(std::io::Error),
+  ImageError(image::ImageError),
+}
+
+impl Error for ImagesError {}
+
+impl Display for ImagesError {
+  fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+    match self {
+      ImagesError::IoError(err) => write!(f, "IO Error: {}", err),
+      ImagesError::ImageError(err) => write!(f, "Image Error: {}", err),
+    }
+  }
+}
+
+impl From<std::io::Error> for ImagesError {
+  fn from(err: std::io::Error) -> Self {
+    ImagesError::IoError(err)
+  }
+}
+
+impl From<image::ImageError> for ImagesError {
+  fn from(err: image::ImageError) -> Self {
+    ImagesError::ImageError(err)
+  }
+}

crates/lib/files/images/src/lib.rs

@@ -23,6 +23,8 @@
 // Re-export image to everyone importing this library
 pub use image;
 
+pub mod encoding;
+pub mod error;
 pub mod image_info;
 pub mod resize_image_file_preserving_aspect;
 pub mod resize_preserving_aspect;

crates/schema/public/api/artcraft_api_defs/src/generate/image/generate_gpt_image_1_text_to_image.rs

@@ -0,0 +1,58 @@
+use serde_derive::{Deserialize, Serialize};
+use tokens::tokens::generic_inference_jobs::InferenceJobToken;
+use tokens::tokens::media_files::MediaFileToken;
+use utoipa::ToSchema;
+
+pub const GPT_IMAGE_1_TEXT_TO_IMAGE_PATH: &str = "/v1/generate/image/gpt_image_1_text_to_image";
+
+#[derive(Serialize, Deserialize, ToSchema)]
+pub struct GenerateGptImage1TextToImageRequest {
+  /// Idempotency token to prevent duplicate requests.
+  pub uuid_idempotency_token: String,
+
+  /// Text prompt to generate the image from.
+  pub prompt: Option<String>,
+
+  /// Size of the image to generate. Default is Square.
+  pub image_size: Option<GenerateGptImage1TextToImageImageSize>,
+
+  /// Number of images to generate. Default is one.
+  pub num_images: Option<GenerateGptImage1TextToImageNumImages>,
+
+  /// Quality of the image to generate. Default is High.
+  pub image_quality: Option<GenerateGptImage1TextToImageImageQuality>,
+}
+
+
+#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
+#[serde(rename_all = "snake_case")]
+pub enum GenerateGptImage1TextToImageImageSize {
+  Square, // 1:1, Default
+  Horizontal,
+  Vertical,
+}
+
+#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
+#[serde(rename_all = "snake_case")]
+pub enum GenerateGptImage1TextToImageNumImages {
+  One, // Default
+  Two,
+  Three,
+  Four,
+}
+
+#[derive(Copy, Clone, Debug, Serialize, Deserialize, ToSchema)]
+#[serde(rename_all = "snake_case")]
+pub enum GenerateGptImage1TextToImageImageQuality {
+  Auto,
+  Low,
+  Medium,
+  High, // Default
+}
+
+
+#[derive(Serialize, Deserialize, ToSchema)]
+pub struct GenerateGptImage1TextToImageResponse {
+  pub success: bool,
+  pub inference_job_token: InferenceJobToken,
+}

crates/schema/public/api/artcraft_api_defs/src/generate/image/mod.rs

@@ -3,4 +3,5 @@ pub mod generate_flux_1_dev_text_to_image;
 pub mod generate_flux_1_schnell_text_to_image;
 pub mod generate_flux_pro_11_text_to_image;
 pub mod generate_flux_pro_11_ultra_text_to_image;
+pub mod generate_gpt_image_1_text_to_image;
 pub mod remove_image_background;

crates/service/web/storyteller_web/src/docs/api_doc.rs

@@ -141,6 +141,11 @@ use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image
 use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image::GenerateFluxPro11UltraTextToImageNumImages;
 use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image::GenerateFluxPro11UltraTextToImageRequest;
 use artcraft_api_defs::generate::image::generate_flux_pro_11_ultra_text_to_image::GenerateFluxPro11UltraTextToImageResponse;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::GenerateGptImage1TextToImageImageQuality;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::GenerateGptImage1TextToImageImageSize;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::GenerateGptImage1TextToImageNumImages;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::GenerateGptImage1TextToImageRequest;
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::GenerateGptImage1TextToImageResponse;
 use artcraft_api_defs::generate::image::remove_image_background::RemoveImageBackgroundRequest;
 use artcraft_api_defs::generate::image::remove_image_background::RemoveImageBackgroundResponse;
 use artcraft_api_defs::generate::object::generate_hunyuan_2_0_image_to_3d::GenerateHunyuan20ImageTo3dRequest;
@@ -237,6 +242,7 @@ use tokens::tokens::zs_voice_datasets::*;
     crate::http_server::endpoints::generate::image::generate_flux_1_dev_text_to_image_handler::generate_flux_1_dev_text_to_image_handler,
     crate::http_server::endpoints::generate::image::generate_flux_1_schnell_text_to_image_handler::generate_flux_1_schnell_text_to_image_handler,
     crate::http_server::endpoints::generate::image::generate_flux_pro_11_text_to_image_handler::generate_flux_pro_11_text_to_image_handler,
+    crate::http_server::endpoints::generate::image::generate_gpt_image_1_text_to_image_handler::generate_gpt_image_1_text_to_image_handler,
     crate::http_server::endpoints::generate::image::generate_flux_pro_11_ultra_text_to_image_handler::generate_flux_pro_11_ultra_text_to_image_handler,
     crate::http_server::endpoints::generate::image::remove_image_background_handler::remove_image_background_handler,
     crate::http_server::endpoints::generate::image::edit::gpt_image_1_edit_image_handler::gpt_image_1_edit_image_handler,
@@ -420,6 +426,11 @@ use tokens::tokens::zs_voice_datasets::*;
     BatchGetUserRatingError,
     BatchGetUserRatingQueryParams,
     BatchGetUserRatingResponse,
+    GenerateGptImage1TextToImageImageQuality,
+    GenerateGptImage1TextToImageImageSize,
+    GenerateGptImage1TextToImageNumImages,
+    GenerateGptImage1TextToImageRequest,
+    GenerateGptImage1TextToImageResponse,
     BatchInferenceJobStatusResponsePayload,
     BatchMediaFileInfo,
     BatchRequestDetailsResponse,

crates/service/web/storyteller_web/src/http_server/endpoints/generate/image/generate_gpt_image_1_text_to_image_handler.rs

@@ -0,0 +1,213 @@
+use std::sync::Arc;
+
+use crate::http_server::common_responses::common_web_error::CommonWebError;
+use crate::http_server::common_responses::media::media_links_builder::MediaLinksBuilder;
+use crate::http_server::endpoints::media_files::helpers::get_media_domain::get_media_domain;
+use crate::http_server::validations::validate_idempotency_token_format::validate_idempotency_token_format;
+use crate::state::server_state::ServerState;
+use actix_web::web::Json;
+use actix_web::{web, HttpRequest};
+use artcraft_api_defs::generate::image::generate_gpt_image_1_text_to_image::{GenerateGptImage1TextToImageImageQuality, GenerateGptImage1TextToImageImageSize, GenerateGptImage1TextToImageNumImages, GenerateGptImage1TextToImageRequest, GenerateGptImage1TextToImageResponse};
+use bucket_paths::legacy::typified_paths::public::media_files::bucket_file_path::MediaFileBucketPath;
+use enums::by_table::prompt_context_items::prompt_context_semantic_type::PromptContextSemanticType;
+use enums::by_table::prompts::prompt_type::PromptType;
+use enums::common::generation_provider::GenerationProvider;
+use enums::common::model_type::ModelType;
+use enums::common::visibility::Visibility;
+use fal_client::creds::open_ai_api_key::OpenAiApiKey;
+use fal_client::requests::webhook::image::enqueue_gpt_image_1_text_to_image_webhook::{enqueue_gpt_image_1_text_to_image_webhook, GptTextToImageByokArgs, GptTextToImageNumImages, GptTextToImageQuality, GptTextToImageSize};
+use http_server_common::request::get_request_ip::get_request_ip;
+use log::{error, info, warn};
+use mysql_queries::queries::generic_inference::fal::insert_generic_inference_job_for_fal_queue::insert_generic_inference_job_for_fal_queue;
+use mysql_queries::queries::generic_inference::fal::insert_generic_inference_job_for_fal_queue::FalCategory;
+use mysql_queries::queries::generic_inference::fal::insert_generic_inference_job_for_fal_queue::InsertGenericInferenceForFalArgs;
+use mysql_queries::queries::idepotency_tokens::insert_idempotency_token::insert_idempotency_token;
+use mysql_queries::queries::media_files::get::batch_get_media_files_by_tokens::{batch_get_media_files_by_tokens, batch_get_media_files_by_tokens_with_connection};
+use mysql_queries::queries::prompt_context_items::insert_batch_prompt_context_items::{insert_batch_prompt_context_items, InsertBatchArgs, PromptContextItem};
+use mysql_queries::queries::prompts::insert_prompt::{insert_prompt, InsertPromptArgs};
+use sqlx::Acquire;
+use utoipa::ToSchema;
+
+/// Gpt Image 1
+#[utoipa::path(
+  post,
+  tag = "Generate Images",
+  path = "/v1/generate/image/gpt_image_1_text_to_image",
+  responses(
+    (status = 200, description = "Success", body = GenerateGptImage1TextToImageResponse),
+  ),
+  params(
+    ("request" = GenerateGptImage1TextToImageRequest, description = "Payload for Request"),
+  )
+)]
+pub async fn generate_gpt_image_1_text_to_image_handler(
+  http_request: HttpRequest,
+  request: Json<GenerateGptImage1TextToImageRequest>,
+  server_state: web::Data<Arc<ServerState>>
+) -> Result<Json<GenerateGptImage1TextToImageResponse>, CommonWebError> {
+  let mut mysql_connection = server_state.mysql_pool
+      .acquire()
+      .await?;
+  
+  let maybe_user_session = server_state
+      .session_checker
+      .maybe_get_user_session_from_connection(&http_request, &mut mysql_connection)
+      .await
+      .map_err(|e| {
+        warn!("Session checker error: {:?}", e);
+        CommonWebError::ServerError
+      })?;
+
+  let maybe_avt_token = server_state
+      .avt_cookie_manager
+      .get_avt_token_from_request(&http_request);
+
+  // TODO: Limit usage for new accounts. Billing, free credits metering, etc.
+
+  //let user_session = match maybe_user_session {
+  //  Some(session) => session,
+  //  None => {
+  //    warn!("not logged in");
+  //    return Err(CommonWebError::NotAuthorized);
+  //  }
+  //};
+
+  if let Err(reason) = validate_idempotency_token_format(&request.uuid_idempotency_token) {
+    return Err(CommonWebError::BadInputWithSimpleMessage(reason));
+  }
+
+  const CAN_SEE_DELETED: bool = false;
+  
+  let media_domain = get_media_domain(&http_request);
+  
+  insert_idempotency_token(&request.uuid_idempotency_token, &mut *mysql_connection)
+      .await
+      .map_err(|err| {
+        error!("Error inserting idempotency token: {:?}", err);
+        CommonWebError::BadInputWithSimpleMessage("repeated idempotency token".to_string())
+      })?;
+
+  info!("Fal webhook URL: {}", server_state.fal.webhook_url);
+
+  let image_size = match request.image_size {
+    Some(GenerateGptImage1TextToImageImageSize::Square) => GptTextToImageSize::Square,
+    Some(GenerateGptImage1TextToImageImageSize::Horizontal) => GptTextToImageSize::Horizontal,
+    Some(GenerateGptImage1TextToImageImageSize::Vertical) => GptTextToImageSize::Vertical,
+    None => GptTextToImageSize::Square, // Default to Square
+  };
+
+  let num_images = match request.num_images {
+    Some(GenerateGptImage1TextToImageNumImages::One) => GptTextToImageNumImages::One,
+    Some(GenerateGptImage1TextToImageNumImages::Two) => GptTextToImageNumImages::Two,
+    Some(GenerateGptImage1TextToImageNumImages::Three) => GptTextToImageNumImages::Three,
+    Some(GenerateGptImage1TextToImageNumImages::Four) => GptTextToImageNumImages::Four,
+    None => GptTextToImageNumImages::One, // Default to One
+  };
+  
+  let quality = match request.image_quality {
+    Some(quality) => match quality {
+      GenerateGptImage1TextToImageImageQuality::Auto => GptTextToImageQuality::Auto,
+      GenerateGptImage1TextToImageImageQuality::Low => GptTextToImageQuality::Low,
+      GenerateGptImage1TextToImageImageQuality::Medium => GptTextToImageQuality::Medium,
+      GenerateGptImage1TextToImageImageQuality::High => GptTextToImageQuality::High,
+    },
+    None => GptTextToImageQuality::High, // Default to High
+  };
+  
+  let openai_api_key = OpenAiApiKey::from_str(&server_state.openai.api_key);
+
+  let args = GptTextToImageByokArgs {
+    prompt: request.prompt.as_deref().unwrap_or(""),
+    webhook_url: &server_state.fal.webhook_url,
+    api_key: &server_state.fal.api_key,
+    num_images,
+    image_size,
+    quality,
+    openai_api_key: &openai_api_key,
+  };
+
+  let fal_result = enqueue_gpt_image_1_text_to_image_webhook(args)
+      .await
+      .map_err(|err| {
+        warn!("Error calling enqueue_gpt_image_1_text_to_image_webhook: {:?}", err);
+        CommonWebError::ServerError
+      })?;
+
+  let external_job_id = fal_result.request_id
+      .ok_or_else(|| {
+        warn!("Fal request_id is None");
+        CommonWebError::ServerError
+      })?;
+
+  info!("Fal request_id: {}", external_job_id);
+
+  let ip_address = get_request_ip(&http_request);
+
+  let mut transaction = mysql_connection
+      .begin()
+      .await
+      .map_err(|err| {
+        error!("Error starting MySQL transaction: {:?}", err);
+        CommonWebError::ServerError
+      })?;
+
+  // NB: Don't fail the job if the query fails.
+  let prompt_result = insert_prompt(InsertPromptArgs {
+    maybe_apriori_prompt_token: None,
+    prompt_type: PromptType::ArtcraftApp,
+    maybe_creator_user_token: maybe_user_session
+        .as_ref()
+        .map(|s| &s.user_token),
+    maybe_model_type: Some(ModelType::GptImage1),
+    maybe_generation_provider: Some(GenerationProvider::Artcraft),
+    maybe_positive_prompt: request.prompt.as_deref(),
+    maybe_negative_prompt: None,
+    maybe_other_args: None,
+    creator_ip_address: &ip_address,
+    mysql_executor: &mut *transaction,
+    phantom: Default::default(),
+  }).await;
+
+  let prompt_token = match prompt_result {
+    Ok(token) => Some(token),
+    Err(err) => {
+      warn!("Error inserting prompt: {:?}", err);
+      None // Don't fail the job if the prompt insertion fails.
+    }
+  };
+  
+  let db_result = insert_generic_inference_job_for_fal_queue(InsertGenericInferenceForFalArgs {
+    uuid_idempotency_token: &request.uuid_idempotency_token,
+    maybe_external_third_party_id: &external_job_id,
+    fal_category: FalCategory::ImageGeneration,
+    maybe_inference_args: None,
+    maybe_prompt_token: prompt_token.as_ref(),
+    maybe_creator_user_token: maybe_user_session.as_ref().map(|s| &s.user_token),
+    maybe_avt_token: maybe_avt_token.as_ref(),
+    creator_ip_address: &ip_address,
+    creator_set_visibility: Visibility::Public,
+    mysql_executor: &mut *transaction,
+    phantom: Default::default(),
+  }).await;
+
+  let job_token = match db_result {
+    Ok(token) => token,
+    Err(err) => {
+      warn!("Error inserting generic inference job for FAL queue: {:?}", err);
+      return Err(CommonWebError::ServerError);
+    }
+  };
+  
+  let _r = transaction
+      .commit()
+      .await
+      .map_err(|err| {
+        error!("Error committing MySQL transaction: {:?}", err);
+        CommonWebError::ServerError
+      })?;
+
+  Ok(Json(GenerateGptImage1TextToImageResponse {
+    success: true,
+    inference_job_token: job_token,
+  }))
+}

crates/service/web/storyteller_web/src/http_server/endpoints/generate/image/mod.rs

@@ -3,4 +3,5 @@ pub mod generate_flux_1_dev_text_to_image_handler;
 pub mod generate_flux_1_schnell_text_to_image_handler;
 pub mod generate_flux_pro_11_text_to_image_handler;
 pub mod generate_flux_pro_11_ultra_text_to_image_handler;
+pub mod generate_gpt_image_1_text_to_image_handler;
 pub mod remove_image_background_handler;

crates/service/web/storyteller_web/src/http_server/endpoints/webhooks/handle_image_payload.rs

@@ -7,8 +7,10 @@ use enums::by_table::media_files::media_file_origin_category::MediaFileOriginCat
 use enums::by_table::media_files::media_file_type::MediaFileType;
 use errors::AnyhowResult;
 use hashing::sha256::sha256_hash_bytes::sha256_hash_bytes;
+use images::encoding::webp_bytes_to_png_bytes::webp_bytes_to_png_bytes;
 use images::image_info::image_info::ImageInfo;
 use log::info;
+use mimetypes::mimetype_info::file_extension::FileExtension;
 use mimetypes::mimetype_info::mimetype_info::MimetypeInfo;
 use mysql_queries::queries::generic_inference::fal::get_inference_job_by_fal_id::FalJobDetails;
 use mysql_queries::queries::media_files::create::insert_builder::media_file_insert_builder::MediaFileInsertBuilder;
@@ -32,18 +34,17 @@ pub async fn handle_image_payload(
   job: &FalJobDetails,
   server_state: &ServerState,
 ) -> AnyhowResult<MediaFileToken> {
-  
   let image_value = payload.get("image")
       .ok_or_else(|| anyhow!("no `image` key in payload"))?;
 
   info!("Fal Image Payload: {:?}", image_value);
-  
+
   let image: FalWebhookImage = serde_json::from_value(image_value.clone())?;
 
   let image_url = image.url
       .as_deref()
       .ok_or_else(|| anyhow!("no `url` in image payload"))?;
-  
+
   //let mime_type = image.content_type
   //    .as_deref()
   //    .ok_or_else(|| anyhow!("no `content_type` in image payload"))?;
@@ -51,10 +52,53 @@ pub async fn handle_image_payload(
   let file_bytes = http_download_url_to_bytes(image_url)
       .await
       .map_err(|e| anyhow!("Failed to download image: {:?}", e))?;
-  
+
   let mimetype_info = MimetypeInfo::get_for_bytes(&file_bytes)
       .ok_or_else(|| anyhow!("Failed to get mimetype info"))?;
-  
+
+
+  info!("File type: {}, extension: {:?}",
+       mimetype_info.mime_type(),
+       mimetype_info.file_extension());
+
+  match mimetype_info.file_extension() {
+    Some(FileExtension::Webp) => {
+      info!("Artcraft can't handle WebP images yet; converting to PNG...");
+
+      let file_bytes = webp_bytes_to_png_bytes(&file_bytes)?;
+
+      let mimetype_info = MimetypeInfo::get_for_bytes(&file_bytes)
+          .ok_or_else(|| anyhow!("Failed to get mimetype info"))?;
+
+      info!("Updated file type: {}, extension: {:?}",
+        mimetype_info.mime_type(),
+        mimetype_info.file_extension());
+
+      upload_single_image_bytes(
+        job,
+        server_state,
+        &file_bytes,
+        mimetype_info,
+      ).await
+    }
+    _ => {
+      upload_single_image_bytes(
+        job,
+        server_state,
+        &file_bytes,
+        mimetype_info,
+      ).await
+    }
+  }
+}
+
+async fn upload_single_image_bytes(
+  job: &FalJobDetails,
+  server_state: &ServerState,
+  file_bytes: &[u8],
+  mimetype_info: MimetypeInfo,
+) -> AnyhowResult<MediaFileToken> {
+
   let mime_type = mimetype_info.mime_type();
 
   let media_file_type = MediaFileType::try_from_mime_type(mime_type)

crates/service/web/storyteller_web/src/http_server/endpoints/webhooks/handle_images_payload.rs

@@ -7,8 +7,10 @@ use enums::by_table::media_files::media_file_origin_category::MediaFileOriginCat
 use enums::by_table::media_files::media_file_type::MediaFileType;
 use errors::AnyhowResult;
 use hashing::sha256::sha256_hash_bytes::sha256_hash_bytes;
+use images::encoding::webp_bytes_to_png_bytes::webp_bytes_to_png_bytes;
 use images::image_info::image_info::ImageInfo;
 use log::info;
+use mimetypes::mimetype_info::file_extension::FileExtension;
 use mimetypes::mimetype_info::mimetype_info::MimetypeInfo;
 use mysql_queries::queries::generic_inference::fal::get_inference_job_by_fal_id::FalJobDetails;
 use mysql_queries::queries::media_files::create::insert_builder::media_file_insert_builder::MediaFileInsertBuilder;
@@ -53,7 +55,7 @@ pub async fn handle_images_payload(
   Ok(maybe_media_token)
 }
 
-pub async fn upload_image(
+async fn upload_image(
   job: &FalJobDetails,
   server_state: &ServerState,
   image: &FalWebhookImage,
@@ -73,11 +75,58 @@ pub async fn upload_image(
   let mimetype_info = MimetypeInfo::get_for_bytes(&file_bytes)
       .ok_or_else(|| anyhow!("Failed to get mimetype info"))?;
 
+  info!("File type: {}, extension: {:?}",
+       mimetype_info.mime_type(),
+       mimetype_info.file_extension());
+
+  match mimetype_info.file_extension() {
+    Some(FileExtension::Webp) => {
+      info!("Artcraft can't handle WebP images yet; converting to PNG...");
+
+      let file_bytes = webp_bytes_to_png_bytes(&file_bytes)?;
+
+      let mimetype_info = MimetypeInfo::get_for_bytes(&file_bytes)
+          .ok_or_else(|| anyhow!("Failed to get mimetype info"))?;
+
+      info!("Updated file type: {}, extension: {:?}",
+        mimetype_info.mime_type(),
+        mimetype_info.file_extension());
+
+      upload_image_bytes(
+        job,
+        server_state,
+        image,
+        &file_bytes,
+        mimetype_info,
+      ).await
+    }
+    _ => {
+      upload_image_bytes(
+        job,
+        server_state,
+        image,
+        &file_bytes,
+        mimetype_info,
+      ).await
+    }
+  }
+}
+
+async fn upload_image_bytes(
+  job: &FalJobDetails,
+  server_state: &ServerState,
+  image: &FalWebhookImage,
+  file_bytes: &[u8],
+  mimetype_info: MimetypeInfo,
+) -> AnyhowResult<MediaFileToken> {
+
   let mime_type = mimetype_info.mime_type();
 
   let media_file_type = MediaFileType::try_from_mime_type(mime_type)
       .ok_or_else(|| anyhow!("Unsupported media file type: {}", mime_type))?;
 
+  info!("MediaFileType: {:?}", media_file_type);
+
   let extension_with_period = mimetype_info.file_extension()
       .map(|ext| ext.extension_with_period())
       .ok_or_else(|| anyhow!("Failed to get file extension from mimetype info"))?;
@@ -117,4 +166,4 @@ pub async fn upload_image(
   info!("Image media file uploaded with token: {}", media_token);
 
   Ok(media_token)
-}
+}

crates/service/web/storyteller_web/src/http_server/routes/add_generate_routes.rs

@@ -1,7 +1,9 @@
+use crate::http_server::endpoints::generate::image::edit::gpt_image_1_edit_image_handler::gpt_image_1_edit_image_handler;
 use crate::http_server::endpoints::generate::image::generate_flux_1_dev_text_to_image_handler::generate_flux_1_dev_text_to_image_handler;
 use crate::http_server::endpoints::generate::image::generate_flux_1_schnell_text_to_image_handler::generate_flux_1_schnell_text_to_image_handler;
 use crate::http_server::endpoints::generate::image::generate_flux_pro_11_text_to_image_handler::generate_flux_pro_11_text_to_image_handler;
 use crate::http_server::endpoints::generate::image::generate_flux_pro_11_ultra_text_to_image_handler::generate_flux_pro_11_ultra_text_to_image_handler;
+use crate::http_server::endpoints::generate::image::generate_gpt_image_1_text_to_image_handler::generate_gpt_image_1_text_to_image_handler;
 use crate::http_server::endpoints::generate::image::remove_image_background_handler::remove_image_background_handler;
 use crate::http_server::endpoints::generate::object::generate_hunyuan_2_0_image_to_3d_handler::generate_hunyuan_2_0_image_to_3d_handler;
 use crate::http_server::endpoints::generate::object::generate_hunyuan_2_1_image_to_3d_handler::generate_hunyuan_2_1_image_to_3d_handler;
@@ -14,7 +16,6 @@ use actix_http::body::MessageBody;
 use actix_service::ServiceFactory;
 use actix_web::dev::{ServiceRequest, ServiceResponse};
 use actix_web::{web, App, Error, HttpResponse};
-use crate::http_server::endpoints::generate::image::edit::gpt_image_1_edit_image_handler::gpt_image_1_edit_image_handler;
 
 pub fn add_generate_routes<T, B> (app: App<T>) -> App<T>
 where
@@ -51,6 +52,10 @@ where
               .route(web::post().to(generate_flux_pro_11_ultra_text_to_image_handler))
               .route(web::head().to(|| HttpResponse::Ok()))
           )
+          .service(web::resource("/gpt_image_1_text_to_image")
+              .route(web::post().to(generate_gpt_image_1_text_to_image_handler))
+              .route(web::head().to(|| HttpResponse::Ok()))
+          )
           .service(web::resource("/remove_background")
               .route(web::post().to(remove_image_background_handler))
               .route(web::head().to(|| HttpResponse::Ok()))

crates/vendor/fal-rs/src/endpoints/fal_ai/gpt_image_1/edit_image/byok/mod.rs

@@ -18,6 +18,22 @@ use crate::prelude::{Deserialize, FalRequest, Serialize};
 }
 */
 
+/// This is the non-edit, text-to-image only endpoint
+#[derive(Debug, Serialize, Deserialize)]
+pub struct GptTextToImageRequest {
+  /// The prompt
+  pub prompt: String,
+
+  pub image_size: String,
+
+  pub num_images: u8,
+
+  pub quality: String,
+
+  pub openai_api_key: String,
+}
+
+/// This is edit-only, image-to-image endpoint
 #[derive(Debug, Serialize, Deserialize)]
 pub struct GptEditImageRequest {
   /// The context images
@@ -39,6 +55,10 @@ pub struct GptEditImageRequest {
 pub struct ImageOutput {
 }
 
+pub fn gpt_text_to_image(params: GptTextToImageRequest) -> FalRequest<GptTextToImageRequest, ImageOutput> {
+  FalRequest::new("fal-ai/gpt-image-1/text-to-image/byok", params)
+}
+
 pub fn gpt_edit_image(params: GptEditImageRequest) -> FalRequest<GptEditImageRequest, ImageOutput> {
   FalRequest::new("fal-ai/gpt-image-1/edit-image/byok", params)
 }

frontend/libs/tauri-api/src/lib/enqueue/EnqueueTextToImage.ts

@@ -17,12 +17,14 @@ export interface EnqueueTextToImageRequest {
   prompt?: string;
   model?: ModelInfo | EnqueueTextToImageModel;
   aspect_ratio?: EnqueueTextToImageSize;
+  number_images?: number;
 }
 
 interface EnqueueTextToImageRawRequest {
   prompt?: string;
   model?: EnqueueTextToImageModel | string; // TODO: Shouldn't allow string
   aspect_ratio?: EnqueueTextToImageSize;
+  number_images?: number;
 }
 
 export enum EnqueueTextToImageModel {
@@ -75,6 +77,10 @@ export const EnqueueTextToImage = async (request: EnqueueTextToImageRequest) : P
     mutableRequest.aspect_ratio = request.aspect_ratio;
   }
 
+  if (!!request.number_images) {
+    mutableRequest.number_images = request.number_images;
+  }
+
   const result = await invoke("enqueue_text_to_image_command", { 
     request: mutableRequest,
   });