Update app.py

app.py

@@ -1,78 +1,69 @@
 # app.py
 # =============
-# This is a complete app.py file for a text generation app using the Qwen/Qwen2.5-Coder-0.5B-Instruct-GGUF model.
-# The app is built using Gradio and runs on a CPU without video memory.
+# This is a complete app.py file for a Gradio app using the meta-llama/Llama-3.2-3B-Instruct model.
+# The app allows users to input a message and receive a response from the model.
 
-# Imports
-# =======
-import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import torch
+# Dependencies
+# =============
+# The following dependencies are required to run this app:
+# - transformers
+# - gradio
+# - torch
+#
+# You can install these dependencies using pip:
+# pip install transformers gradio torch
 
-# Constants
-# =========
-MODEL_NAME = "Qwen/Qwen2.5-Coder-0.5B-Instruct-GGUF"
-DEVICE = "cpu"  # Ensure the model runs on CPU
+import torch
+from transformers import pipeline
+import gradio as gr  # Import gradio
 
-# Load Model and Tokenizer
-# ========================
-def load_model_and_tokenizer():
-    """
-    Load the model and tokenizer from Hugging Face.
-    """
-    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
-    model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype=torch.bfloat16, device_map=DEVICE)
-    return tokenizer, model
+# Load the model and tokenizer
+model_id = "meta-llama/Llama-3.2-3B-Instruct"
+device = "cpu"  # Use CPU for inference
 
-tokenizer, model = load_model_and_tokenizer()
+# Initialize the pipeline
+pipe = pipeline(
+    "text-generation",
+    model=model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
 
-# Generate Text
-# =============
-def generate_text(prompt, max_length=100):
+def generate_response(prompt):
     """
-    Generate text based on the given prompt.
+    Generate a response from the model based on the given prompt.
     
     Args:
-        prompt (str): The input prompt for text generation.
-        max_length (int): The maximum length of the generated text.
+        prompt (str): The input message from the user.
     
     Returns:
-        str: The generated text.
+        str: The generated response from the model.
     """
-    inputs = tokenizer(prompt, return_tensors="pt").to(DEVICE)
-    outputs = model.generate(inputs.input_ids, max_length=max_length, num_return_sequences=1)
-    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return generated_text
+    messages = [
+        {"role": "system", "content": "You are a helpful assistant!"},
+        {"role": "user", "content": prompt},
+    ]
+    outputs = pipe(
+        messages,
+        max_new_tokens=256,
+    )
+    return outputs[0]["generated_text"][-1]
 
-# Gradio Interface
-# =================
+# Define the Gradio interface
 def gradio_interface():
     """
-    Create and launch the Gradio interface.
+    Define the Gradio interface for the app.
     """
     iface = gr.Interface(
-        fn=generate_text,
-        inputs=[
-            gr.inputs.Textbox(lines=2, placeholder="Enter your prompt here..."),
-            gr.inputs.Slider(minimum=50, maximum=500, step=10, default=100, label="Max Length")
-        ],
+        fn=generate_response,
+        inputs=gr.inputs.Textbox(lines=2, placeholder="Enter your message here..."),
         outputs="text",
-        title="Qwen2.5-Coder-0.5B-Instruct-GGUF Text Generation",
-        description="Generate text using the Qwen2.5-Coder-0.5B-Instruct-GGUF model."
+        title="Llama-3.2-3B-Instruct Chatbot",
+        description="Chat with the Llama-3.2-3B-Instruct model. Enter your message and get a response!",
     )
-    iface.launch()
+    return iface
 
-# Main
-# ====
+# Launch the Gradio app
 if __name__ == "__main__":
-    gradio_interface()
-
-# Dependencies
-# =============
-# The following dependencies are required to run this app:
-# - transformers
-# - gradio
-# - torch
-#
-# You can install these dependencies using pip:
-# pip install transformers gradio torch
+    iface = gradio_interface()
+    iface.launch()