optimized for new video

app.py

@@ -1,25 +1,21 @@
 from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 import gradio as gr
 
-# Define the model name
-model_name = "tiiuae/Falcon3-10B-Instruct"
+model_name = "tiiuae/falcon3-10b-instruct"
 
-# Load the tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
-    torch_dtype="auto",
-    device_map="auto"
+    torch_dtype=torch.float16,  # Use FP16 precision
+    device_map="auto"          # Automatically maps to GPU
 )
 
-# Define a simple chat function
 def chat(input_text):
-    inputs = tokenizer(input_text, return_tensors="pt")
+    inputs = tokenizer(input_text, return_tensors="pt").to('cuda')
     outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True)
-    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-# Create a Gradio interface
 interface = gr.Interface(
     fn=chat,
     inputs=gr.Textbox(label="Your Message"),
@@ -27,5 +23,4 @@ interface = gr.Interface(
     title="Falcon-10B-Instruct Chat"
 )
 
-# Launch the app
 interface.launch()