Changing to run LLama 3.2

app.py

@@ -1,11 +1,13 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load Llama 3.2-3B-Instruct model locally
+model_name = "meta-llama/Llama-3.2-3B-Instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name, torch_dtype=torch.float16, device_map="auto"
+)
 
 def respond(
     message,
@@ -15,8 +17,8 @@ def respond(
     temperature,
     top_p,
 ):
+    # Format the conversation history
     messages = [{"role": "system", "content": system_message}]
-
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
@@ -24,25 +26,20 @@ def respond(
             messages.append({"role": "assistant", "content": val[1]})
 
     messages.append({"role": "user", "content": message})
+    prompt = "\n".join([msg["content"] for msg in messages])
 
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    # Tokenize and generate response
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Gradio ChatInterface with controls for temperature, tokens, etc.
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -59,6 +56,5 @@ demo = gr.ChatInterface(
     ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()