Update app.py

app.py

@@ -5,18 +5,43 @@ import torch
 # Load Model dan Tokenizer
 model_name = "meta-llama/Llama-3.3-70B-Instruct"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype=torch.float16)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    device_map="auto",
+    torch_dtype=torch.float16
+)
 
 # Fungsi Chatbot
 def chat(input_text):
     # Tokenisasi input
-    inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
+    inputs = tokenizer(
+        input_text,
+        return_tensors="pt",
+        truncation=True,  # Potong jika terlalu panjang
+        max_length=512    # Batasi panjang input
+    ).to("cuda")
+    
     # Hasilkan output menggunakan model
-    outputs = model.generate(**inputs, max_length=1000, top_p=0.9, temperature=0.7)
+    outputs = model.generate(
+        **inputs,
+        max_length=200,       # Panjang maksimal output
+        top_p=0.9,            # Sampling berdasarkan probabilitas kumulatif
+        temperature=0.7,      # Kontrol kreativitas output
+        num_return_sequences=1 # Jumlah respons yang dihasilkan
+    )
+    
     # Decode output menjadi teks
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
     return response
 
 # Membuat antarmuka Gradio
-iface = gr.Interface(fn=chat, inputs="text", outputs="text", live=True)
+iface = gr.Interface(
+    fn=chat,
+    inputs="text",
+    outputs="text",
+    title="Llama Chatbot",
+    description="Chatbot AI menggunakan model Llama-3.3-70B-Instruct."
+)
+
+# Meluncurkan Gradio
 iface.launch()