Update app.py

GPT-4 "mixer" added to HF Client completion for testing of mixed model output.

app.py

@@ -1,11 +1,12 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import openai
 
-"""
-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")
+# Initialize Hugging Face client
+hf_client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+# OpenAI GPT-4 API key
+openai.api_key = "sk-proj-vA5kIjVnIcoxmcsgTEfBT3BlbkFJ2aD0n6zIUcYzpcu48QHK"
 
 def respond(
     message,
@@ -14,6 +15,7 @@ def respond(
     max_tokens,
     temperature,
     top_p,
+    model_choice,
 ):
     messages = [{"role": "system", "content": system_message}]
 
@@ -27,21 +29,31 @@ def respond(
 
     response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+    if model_choice == "Hugging Face Model":
+        for message in hf_client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            token = message.choices[0].delta.content
+            response += token
+            yield response
+    elif model_choice == "OpenAI GPT-4":
+        response_openai = openai.ChatCompletion.create(
+            model="gpt-4",
+            messages=messages,
+            max_tokens=max_tokens,
+            temperature=temperature,
+            top_p=top_p,
+            stream=True
+        )
+        for message in response_openai:
+            response += message['choices'][0]['delta'].get('content', '')
+            yield response
+
+# Create the Gradio interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -55,9 +67,15 @@ demo = gr.ChatInterface(
             step=0.05,
             label="Top-p (nucleus sampling)",
         ),
+        gr.Radio(
+            choices=["Hugging Face Model", "OpenAI GPT-4"],
+            value="Hugging Face Model",
+            label="Choose Model"
+        )
     ],
+    title="GPT-4 vs Hugging Face Model Comparison",
+    description="Compare responses between a Hugging Face model and OpenAI's GPT-4."
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()