character tokens in omni api (#1333)

crates/api_clients/artcraft/artcraft_api_defs/src/omni_gen/cost_and_generate_requests/omni_gen_video_cost_and_generate_request.rs

@@ -5,6 +5,7 @@ use enums::common::generation::common_aspect_ratio::CommonAspectRatio;
 use enums::common::generation::common_quality::CommonQuality;
 use enums::common::generation::common_resolution::CommonResolution;
 use enums::common::generation::common_video_model::CommonVideoModel;
+use tokens::tokens::characters::CharacterToken;
 use tokens::tokens::media_files::MediaFileToken;
 
 /// Shared request body for both the video cost estimate and video generation endpoints.
@@ -39,6 +40,10 @@ pub struct OmniGenVideoCostAndGenerateRequest {
   /// Reference audio (optional).
   pub reference_audio_media_tokens: Option<Vec<MediaFileToken>>,
 
+  /// Optional character tokens to reference in the prompt.
+  /// Characters are referenced in prompts as @CharacterName.
+  pub reference_character_tokens: Option<Vec<CharacterToken>>,
+
   /// The resolution to use.
   pub resolution: Option<CommonResolution>,
 

crates/api_clients/artcraft/artcraft_api_defs/src/omni_gen/models/omni_gen_video_models.rs

@@ -1,10 +1,13 @@
-use serde_derive::{Deserialize, Serialize};
-use utoipa::{IntoParams, ToSchema};
 use enums::common::generation::common_aspect_ratio::CommonAspectRatio;
 use enums::common::generation::common_quality::CommonQuality;
 use enums::common::generation::common_resolution::CommonResolution;
 use enums::common::generation::common_video_model::CommonVideoModel;
 use enums::common::generation_provider::GenerationProvider;
+use serde_derive::{Deserialize, Serialize};
+use utoipa::{IntoParams, ToSchema};
+
+/// Video model to default to if none is specified
+const DEFAULT_VIDEO_MODEL : CommonVideoModel = CommonVideoModel::Seedance2p0;
 
 /// Query string parameters for the video models endpoint.
 #[derive(Deserialize, IntoParams, ToSchema)]
@@ -110,6 +113,12 @@ pub struct OmniGenVideoModelDetails {
   #[serde(skip_serializing_if = "Option::is_none")]
   pub audio_references_max_total_duration_seconds: Option<u16>,
 
+  #[serde(skip_serializing_if = "Option::is_none")]
+  pub character_references_supported: Option<bool>,
+
+  #[serde(skip_serializing_if = "Option::is_none")]
+  pub character_references_max: Option<u16>,
+
   #[serde(skip_serializing_if = "Option::is_none")]
   pub show_generate_with_sound_toggle: Option<bool>,
 
@@ -159,7 +168,7 @@ pub struct OmniGenVideoModelDetails {
 impl Default for OmniGenVideoModelDetails {
   fn default() -> Self {
     Self {
-      model: CommonVideoModel::Seedance2p0,
+      model: DEFAULT_VIDEO_MODEL,
       full_name: None,
       text_prompt_supported: None,
       text_prompt_max_length: None,
@@ -176,6 +185,8 @@ impl Default for OmniGenVideoModelDetails {
       audio_references_supported: None,
       audio_references_max: None,
       audio_references_max_total_duration_seconds: None,
+      character_references_supported: None,
+      character_references_max: None,
       show_generate_with_sound_toggle: None,
       aspect_ratio_options: None,
       aspect_ratio_default: None,

crates/service/web/storyteller_web/src/configs/omni_gen/video_models.rs

@@ -61,6 +61,8 @@ fn build_omni_gen_video_models() -> Vec<OmniGenVideoModelDetails> {
     video_references_supported: Some(true),
     video_references_max: Some(3),
     video_references_max_total_duration_seconds: Some(15),
+    character_references_supported: Some(true),
+    character_references_max: Some(9),
     aspect_ratio_options: Some(vec![
       CommonAspectRatio::WideSixteenByNine,
       CommonAspectRatio::WideFourByThree,