Update app.py

app.py

@@ -1,18 +1,18 @@
 import os
 import gradio as gr
-from huggingface_hub import login, InferenceClient
+from huggingface_hub import login
+from huggingface_hub import InferenceClient
 import spaces
 
-# Authenticate with Hugging Face API key
+# Retrieve API key and authenticate
 api_key = os.getenv("LLAMA")
 login(api_key)
 
-# Initialize InferenceClients for multiple models
-client1 = InferenceClient("meta-llama/Llama-3.1-70B-Instruct")
-client2 = InferenceClient("bigscience/bloom")
+# Initialize InferenceClient for the Llama model
+client = InferenceClient("meta-llama/Llama-3.1-70B-Instruct")
 
 @spaces.GPU
-def compare_models(
+def respond(
     message,
     history: list[dict],
     system_message,
@@ -22,26 +22,17 @@ def compare_models(
 ):
     # Start with the system message
     messages = [{"role": "system", "content": system_message}]
-    messages += history  # Add conversation history
-    messages.append({"role": "user", "content": message})  # Add user message
 
-    # Fetch responses from both models
-    response1 = ""
-    response2 = ""
+    # Add the conversation history
+    messages += history
 
-    # Stream responses for Model 1
-    for message in client1.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-        response1 += token
+    # Add the latest user message
+    messages.append({"role": "user", "content": message})
+
+    response = ""
 
-    # Stream responses for Model 2
-    for message in client2.chat_completion(
+    # Send the conversation to the model and stream the response
+    for message in client.chat_completion(
         messages,
         max_tokens=max_tokens,
         stream=True,
@@ -49,73 +40,29 @@ def compare_models(
         top_p=top_p,
     ):
         token = message.choices[0].delta.content
-        response2 += token
-
-    # Return responses side-by-side
-    return response1, response2
-
-def handle_vote(vote, current_votes):
-    """Handle user votes."""
-    current_votes[vote] += 1
-    return f"Model 1: {current_votes['model1']} votes | Model 2: {current_votes['model2']} votes"
-
-# Initialize voting state
-votes = {"model1": 0, "model2": 0}
-
-# Create Gradio interface
-with gr.Blocks() as demo:
-    gr.Markdown("# AI Model Comparison Tool")
-    with gr.Row():
-        system_message = gr.Textbox(
-            value="You are a helpful assistant specializing in tech-related topics.",
-            label="System message",
-        )
-        max_tokens = gr.Slider(
-            minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"
-        )
-        temperature = gr.Slider(
-            minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"
-        )
-        top_p = gr.Slider(
-            minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"
-        )
-
-    with gr.Row():
-        message = gr.Textbox(label="Enter your message")
-
-    compare_btn = gr.Button("Compare Models")
-
-    with gr.Row():
-        response1 = gr.Textbox(label="Response from Model 1")
-        response2 = gr.Textbox(label="Response from Model 2")
-
-    with gr.Row():
-        vote_model1 = gr.Button("Vote for Model 1")
-        vote_model2 = gr.Button("Vote for Model 2")
-
-    vote_status = gr.Textbox(
-        value=f"Model 1: {votes['model1']} votes | Model 2: {votes['model2']} votes",
-        label="Voting Results",
-    )
-
-    # Link components
-    compare_btn.click(
-        compare_models,
-        inputs=[message, [], system_message, max_tokens, temperature, top_p],
-        outputs=[response1, response2],
-    )
-
-    vote_model1.click(
-        handle_vote,
-        inputs=["model1", votes],
-        outputs=vote_status,
-    )
-
-    vote_model2.click(
-        handle_vote,
-        inputs=["model2", votes],
-        outputs=vote_status,
-    )
+        response += token
+        yield response
+
+# Initialize the Gradio ChatInterface with the new format
+demo = gr.ChatInterface(
+    respond,
+    type="messages",  # Use the OpenAI-style format
+    additional_inputs=[
+        gr.Textbox(
+            value="You are a helpful Customer Support assistant that specializes in the low-code software company: 'Plant an App' and tech-related topics.",
+            label="System message"
+        ),
+        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.95,
+            step=0.05,
+            label="Top-p (nucleus sampling)"
+        ),
+    ],
+)
 
 if __name__ == "__main__":
     demo.launch()