diff --git a/examples/repl.ts b/examples/repl.ts
index 129678e..ef106eb 100644
--- a/examples/repl.ts
+++ b/examples/repl.ts
@@ -4,7 +4,7 @@ import Polyfact from "../lib/index";
 
 async function main() {
     const { Chat } = Polyfact.exec();
-    const chat = new Chat({ autoMemory: true, provider: "openai" });
+    const chat = new Chat({ autoMemory: true, provider: "openai", model: "gpt-3.5-turbo" });
 
     const rl = readline.createInterface({ input, output });
 
diff --git a/lib/chats/index.ts b/lib/chats/index.ts
index ab53cea..7881f3c 100644
--- a/lib/chats/index.ts
+++ b/lib/chats/index.ts
@@ -47,6 +47,7 @@ export async function createChat(
 
 type ChatOptions = {
     provider?: "openai" | "cohere" | "llama";
+    model?: string;
     systemPrompt?: string;
     autoMemory?: boolean;
 };
@@ -56,6 +57,8 @@ export class Chat {
 
     provider: "openai" | "cohere" | "llama";
 
+    model?: string;
+
     clientOptions: Promise<ClientOptions>;
 
     autoMemory?: Promise<Memory>;
@@ -64,6 +67,7 @@ export class Chat {
         this.clientOptions = defaultOptions(clientOptions);
         this.chatId = createChat(options.systemPrompt, this.clientOptions);
         this.provider = options.provider || "openai";
+        this.model = options.model;
         if (options.autoMemory) {
             this.autoMemory = this.clientOptions.then((co) => new Memory(co));
         }
@@ -81,7 +85,7 @@ export class Chat {
 
         const result = await generateWithTokenUsage(
             message,
-            { provider: this.provider, ...options, chatId },
+            { provider: this.provider, model: this.model, ...options, chatId },
             this.clientOptions,
         );
 
@@ -154,7 +158,7 @@ export class Chat {
 
             const result = generateStream(
                 message,
-                { provider: this.provider, ...options, chatId },
+                { provider: this.provider, model: this.model, ...options, chatId },
                 await this.clientOptions,
             );
 
diff --git a/lib/generate.ts b/lib/generate.ts
index a6d6456..f6261cd 100644
--- a/lib/generate.ts
+++ b/lib/generate.ts
@@ -22,6 +22,7 @@ const GenerationAPIResponse = t.intersection([Required, PartialResultType]);
 
 export type GenerationOptions = {
     provider?: "openai" | "cohere" | "llama";
+    model?: string;
     chatId?: string;
     memory?: Memory;
     memoryId?: string;
@@ -63,11 +64,13 @@ export async function generateWithTokenUsage(
         // eslint-disable-next-line camelcase
         chat_id?: string;
         provider: GenerationOptions["provider"];
+        model?: string;
         stop: GenerationOptions["stop"];
         infos: boolean;
     } = {
         task,
         provider: options?.provider || "openai",
+        model: options.model,
         memory_id: (await options?.memory?.memoryId) || options?.memoryId || "",
         chat_id: options?.chatId || "",
         stop: options?.stop || [],
@@ -140,11 +143,13 @@ function stream(
             // eslint-disable-next-line camelcase
             chat_id?: string;
             provider: GenerationOptions["provider"];
+            model?: string;
             stop: GenerationOptions["stop"];
             infos?: boolean;
         } = {
             task,
             provider: options?.provider || "openai",
+            model: options?.model,
             memory_id: (await options?.memory?.memoryId) || options?.memoryId || "",
             chat_id: options?.chatId || "",
             stop: options?.stop || [],