Create app.py

app.py

@@ -0,0 +1,23 @@
+import gradio as gr
+from huggingface_hub import InferenceClient
+
+client = InferenceClient(model="https://a17d18bb-c94d-4b50-b78f-14306508c584.job.console.elementai.com",
+                         headers={"Authorization":"Bearer j1__hysJQB-F6HXCBqJmmQ:r099ldtg2MtRBdHT6hEeAZndDOj3W68fs58yWNMZJ6M"})
+
+def inference(message, history):
+    partial_message = ""
+    for token in client.text_generation(message, max_new_tokens=500, stream=True):
+        partial_message += token
+        yield partial_message
+
+gr.ChatInterface(
+    inference,
+    chatbot=gr.Chatbot(height=700),
+    textbox=gr.Textbox(placeholder="Prompt CodeLlama model", container=False, scale=2),
+    description="This is the CodeLLaMA 34b-Instruct-hf model. Note that this \"chat\" does not keep context. Each message is a separate prompt.",
+    title="ATG 🤝 TGI",
+    examples=["write a Python function to add 2 numbers", "write a Javascript function that logs out the current time"],
+    retry_btn="Retry",
+    undo_btn="Undo",
+    clear_btn="Clear",
+).queue().launch()