diff --git a/README.md b/README.md
index 98a85be..25a3a28 100644
--- a/README.md
+++ b/README.md
@@ -47,7 +47,7 @@ You can now use this module in any renderer. By default, `@electron/llm` auto-in
 ```
 // First, load the model
 await window.electronAi.create({
-  modelPath: "/full/path/to/model.gguf"
+  modelAlias: "Meta-Llama-3-8B-Instruct.Q4_K_M.gguf"
 })
 
 // Then, talk to it
diff --git a/__tests__/preload.test.ts b/__tests__/preload.test.ts
index de8fdd9..9d093b9 100644
--- a/__tests__/preload.test.ts
+++ b/__tests__/preload.test.ts
@@ -59,6 +59,16 @@ describe('Preload Interface', () => {
     );
   });
 
+  it('prompt should invoke without options', async () => {
+    const input = 'Test prompt';
+    await (globalThis as any).electronAi.prompt(input);
+    expect(ipcRenderer.invoke).toHaveBeenCalledWith(
+      IpcRendererMessage.ELECTRON_LLM_PROMPT,
+      input,
+      undefined,
+    );
+  });
+
   it('promptStreaming should invoke with correct params', async () => {
     const input = 'Test prompt for streaming';
 
diff --git a/src/interfaces.ts b/src/interfaces.ts
index 29116b2..143193b 100644
--- a/src/interfaces.ts
+++ b/src/interfaces.ts
@@ -50,7 +50,7 @@ export interface InternalLanguageModelPromptOptions
 export type AiProcessModelCreateData = InternalLanguageModelCreateOptions;
 
 export interface AiProcessSendPromptData {
-  options: LanguageModelPromptOptions;
+  options?: LanguageModelPromptOptions;
   stream?: boolean;
   input: string;
 }
diff --git a/src/main/register-ai-handlers.ts b/src/main/register-ai-handlers.ts
index bfe12c8..2e151f5 100644
--- a/src/main/register-ai-handlers.ts
+++ b/src/main/register-ai-handlers.ts
@@ -99,7 +99,7 @@ export function registerAiHandlers({
 
   ipcMain.handle(
     IpcRendererMessage.ELECTRON_LLM_PROMPT,
-    async (_event, input: string, options: LanguageModelPromptOptions) => {
+    async (_event, input: string, options?: LanguageModelPromptOptions) => {
       if (!aiProcess) {
         throw new Error(
           'AI model process not started. Please do so with `electronAi.create()`',
@@ -132,7 +132,7 @@ export function registerAiHandlers({
       const timeoutPromise = new Promise((_, reject) => {
         setTimeout(
           () => reject(new Error('Prompt response timed out.')),
-          options.timeout || 20000,
+          options?.timeout || 20000,
         );
       });
 
@@ -142,7 +142,7 @@ export function registerAiHandlers({
 
   ipcMain.on(
     IpcRendererMessage.ELECTRON_LLM_PROMPT_STREAMING_REQUEST,
-    (event, input: string, options: LanguageModelPromptOptions) => {
+    (event, input: string, options?: LanguageModelPromptOptions) => {
       if (!aiProcess) {
         event.sender.send(
           'ELECTRON_LLM_PROMPT_STREAMING_ERROR',
diff --git a/src/utility/call-ai-model-entry-point.ts b/src/utility/call-ai-model-entry-point.ts
index 17bc110..4beec67 100644
--- a/src/utility/call-ai-model-entry-point.ts
+++ b/src/utility/call-ai-model-entry-point.ts
@@ -48,9 +48,9 @@ async function generateResponse(message: PromptMessage) {
     return;
   }
 
-  const options = abortSignalManager.getWithSignalFromPromptOptions(
-    data.options,
-  );
+  const options =
+    data.options &&
+    abortSignalManager.getWithSignalFromPromptOptions(data.options);
 
   try {
     // Format the prompt payload correctly for the language model