diff --git a/crates/api_clients/artcraft/artcraft_api_defs/src/prompts/batch_get_prompts.rs b/crates/api_clients/artcraft/artcraft_api_defs/src/prompts/batch_get_prompts.rs
index 10b7e5638a..5ab9aa7487 100644
--- a/crates/api_clients/artcraft/artcraft_api_defs/src/prompts/batch_get_prompts.rs
+++ b/crates/api_clients/artcraft/artcraft_api_defs/src/prompts/batch_get_prompts.rs
@@ -7,15 +7,18 @@ use enums::common::generation::common_model_class::CommonModelClass;
 use enums::common::generation::common_model_type::CommonModelType;
 use enums::common::generation::common_resolution::CommonResolution;
 use enums::common::generation_provider::GenerationProvider;
+use std::collections::HashSet;
+
 use serde_derive::{Deserialize, Serialize};
 use tokens::tokens::prompts::PromptToken;
-use utoipa::ToSchema;
+use utoipa::{IntoParams, ToSchema};
 
 pub const BATCH_GET_PROMPTS_PATH: &str = "/v1/prompt/batch";
 
-#[derive(Deserialize, ToSchema)]
-pub struct BatchGetPromptsRequest {
-  pub tokens: Vec<PromptToken>,
+#[derive(Deserialize, ToSchema, IntoParams)]
+pub struct BatchGetPromptsQuery {
+  /// Prompt tokens to look up.
+  pub tokens: HashSet<String>,
 }
 
 #[derive(Serialize, ToSchema)]
diff --git a/crates/service/web/storyteller_web/src/docs/api_doc.rs b/crates/service/web/storyteller_web/src/docs/api_doc.rs
index 348aadafd9..35210bc1d7 100644
--- a/crates/service/web/storyteller_web/src/docs/api_doc.rs
+++ b/crates/service/web/storyteller_web/src/docs/api_doc.rs
@@ -625,7 +625,7 @@ use crate::http_server::endpoints::media_files::list::list_batch_generated_redux
     BatchGetInferenceJobStatusQueryParams,
     BatchGetInferenceJobStatusSuccessResponse,
     BatchGetMediaFilesModelInfo,
-    BatchGetPromptsRequest,
+    BatchGetPromptsQuery,
     BatchGetPromptsResponse,
     BatchPromptInfo,
     CreatePromptRequest,
diff --git a/crates/service/web/storyteller_web/src/http_server/endpoints/prompts/batch_get_prompts_handler.rs b/crates/service/web/storyteller_web/src/http_server/endpoints/prompts/batch_get_prompts_handler.rs
index e5d45a35fc..2626fb02b5 100644
--- a/crates/service/web/storyteller_web/src/http_server/endpoints/prompts/batch_get_prompts_handler.rs
+++ b/crates/service/web/storyteller_web/src/http_server/endpoints/prompts/batch_get_prompts_handler.rs
@@ -3,8 +3,9 @@ use std::sync::Arc;
 
 use actix_web::web::Json;
 use actix_web::{web, HttpRequest};
+use actix_web_lab::extract::Query;
 use artcraft_api_defs::prompts::batch_get_prompts::{
-  BatchGetPromptsRequest, BatchGetPromptsResponse, BatchPromptInfo,
+  BatchGetPromptsQuery, BatchGetPromptsResponse, BatchPromptInfo,
 };
 use artcraft_api_defs::prompts::get_prompt::GetPromptImageContextItem;
 use bucket_paths::legacy::typified_paths::public::media_files::bucket_file_path::MediaFileBucketPath;
@@ -22,11 +23,13 @@ use crate::state::server_state::ServerState;
 const MAX_BATCH_SIZE: usize = 100;
 
 /// Batch get details on multiple prompts.
+///
+/// Called, eg. {{api_host}}/v1/prompt/batch?tokens={val}&tokens={val}&tokens=...
 #[utoipa::path(
-  post,
+  get,
   tag = "Prompts",
   path = "/v1/prompt/batch",
-  request_body = BatchGetPromptsRequest,
+  params(BatchGetPromptsQuery),
   responses(
     (status = 200, description = "Found", body = BatchGetPromptsResponse),
     (status = 400, description = "Bad request"),
@@ -35,22 +38,19 @@ const MAX_BATCH_SIZE: usize = 100;
 )]
 pub async fn batch_get_prompts_handler(
   http_request: HttpRequest,
-  body: Json<BatchGetPromptsRequest>,
+  query: Query<BatchGetPromptsQuery>,
   server_state: web::Data<Arc<ServerState>>,
 ) -> Result<Json<BatchGetPromptsResponse>, AdvancedCommonWebError> {
-  let request = body.into_inner();
-
-  if request.tokens.len() > MAX_BATCH_SIZE {
+  if query.tokens.len() > MAX_BATCH_SIZE {
     return Err(AdvancedCommonWebError::BadInputWithSimpleMessage(
-      format!("tokens must contain between 0 and {} items", MAX_BATCH_SIZE),
+      format!("tokens must contain at most {} items", MAX_BATCH_SIZE),
     ));
   }
 
   // Deduplicate and trim whitespace
-  let unique_tokens: Vec<PromptToken> = request
-    .tokens
-    .into_iter()
-    .map(|t| PromptToken::new_from_str(t.as_str().trim()))
+  let unique_tokens: Vec<PromptToken> = query.tokens
+    .iter()
+    .map(|t| PromptToken::new_from_str(t.trim()))
     .filter(|t| !t.as_str().is_empty())
     .collect::<HashSet<_>>()
     .into_iter()
diff --git a/crates/service/web/storyteller_web/src/http_server/routes/application_routes/prompts_routes.rs b/crates/service/web/storyteller_web/src/http_server/routes/application_routes/prompts_routes.rs
index abfc69b8e7..fa34a7dbfd 100644
--- a/crates/service/web/storyteller_web/src/http_server/routes/application_routes/prompts_routes.rs
+++ b/crates/service/web/storyteller_web/src/http_server/routes/application_routes/prompts_routes.rs
@@ -22,7 +22,7 @@ where
   RouteBuilder::from_app(app)
       // NB: This poor RouteBuilder utility requires that POST comes first, otherwise the GET glob
       // will capture it and force 504 Method Not Allowed for all POSTs.
-      .add_post("/v1/prompt/batch", batch_get_prompts_handler)
+      .add_get("/v1/prompt/batch", batch_get_prompts_handler)
       .add_post("/v1/prompts/create", create_prompt_handler)
       .add_get("/v1/prompts/{token}", get_prompt_handler)
       .into_app()