DANIEL PLSSS ADD THIS

LLaMa backup

@@ -0,0 +1,41 @@
+import os
+import gradio as gr
+from huggingface_hub import InferenceClient
+
+# Initialize the Hugging Face Inference Client
+# It automatically picks up the system HF_TOKEN if needed, or runs anonymously for public models
+client = InferenceClient()
+
+def predict(message, history):
+    # Format the entire chat history for an LLM instruction model
+    messages = []
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    
+    # Append the newest message
+    messages.append({"role": "user", "content": message})
+    
+    # Request a streaming response from a powerful, free-tier chat model
+    response = ""
+    for token in client.chat_completion(
+        model="meta-llama/Llama-3.2-3B-Instruct",
+        messages=messages,
+        max_tokens=500,
+        stream=True
+    ):
+        token_text = token.choices[0].delta.content
+        if token_text:
+            response += token_text
+            yield response
+
+# Build a beautiful, responsive Gradio ChatInterface
+demo = gr.ChatInterface(
+    fn=predict,
+    title="My Custom Chatbot",
+    description="Ask me anything! Powered by Llama 3.2 via Hugging Face Inference.",
+    examples=["Explain quantum computing simply.", "Write a poem about a coding error."]
+)
+
+if __name__ == "__main__":
+    demo.launch()