Create app.py

app.py

@@ -0,0 +1,29 @@
+import os
+import gradio as gr
+from huggingface_hub import InferenceClient
+
+MODEL_ID = "meta-llama/Meta-Llama-3.1-8B-Instruct"
+SYSTEM_PROMPT = "You are a helpful assistant."
+
+client = InferenceClient(model=MODEL_ID, token=os.getenv("HF_TOKEN"))
+
+def respond(message, history):
+    # history is OpenAI-style messages when using type="messages"
+    messages = [{"role": "system", "content": SYSTEM_PROMPT}]
+    messages.extend(history)
+    messages.append({"role": "user", "content": message})
+
+    out = client.chat_completion(
+        messages=messages,
+        max_tokens=400,
+        temperature=0.7,
+    )
+    return out.choices[0].message.content
+
+demo = gr.ChatInterface(
+    fn=respond,
+    type="messages",
+    title="Llama 3.1 Chatbot (HF Space)",
+)
+
+demo.launch()