Update app.py

app.py

@@ -1,21 +1,18 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-import os
 
-# Initialize the InferenceClient
-client = InferenceClient("HuggingFaceH4/starchat2-15b-v0.1")
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 def respond(
     message,
-    history,
+    history: list[tuple[str, str]],
     system_message,
     max_tokens,
     temperature,
     top_p,
-    stream,
-    additional_context: str = "",  # Added to include additional context learned from the file
+    stream: bool,
 ):
-    messages = [{"role": "system", "content": system_message + "\n" + additional_context}]
+    messages = [{"role": "system", "content": system_message}]
 
     for val in history:
         if val[0]:
@@ -28,11 +25,11 @@ def respond(
     response = ""
 
     if stream:
-        # Streaming mode: yield response as it comes in
+        # Streaming mode: yield response piece by piece as it comes in
         for message in client.chat_completion(
             messages,
             max_tokens=max_tokens,
-            stream=stream,
+            stream=True,  # Ensures streaming is enabled
             temperature=temperature,
             top_p=top_p,
         ):
@@ -41,59 +38,31 @@ def respond(
             yield response
     else:
         # Non-streaming mode: return the full response at once
-        response = client.chat_completion(
+        result = client.chat_completion(
             messages,
             max_tokens=max_tokens,
             stream=False,  # Disable streaming for full response
             temperature=temperature,
             top_p=top_p,
         )
-        yield response.choices[0].message.content  # Return the full response
+        yield result.choices[0].message.content  # Return full response
 
-# Function to simulate learning from links.txt
-def learn_from_file():
-    if os.path.exists("links.txt"):
-        with open("links.txt", "r") as file:
-            links = file.readlines()
-        # Join the links into a single string to be used as context for the AI
-        additional_context = "\n".join([f"Learned from link: {link.strip()}" for link in links])
-        return additional_context
-    else:
-        return "No file found to learn from."
-
-# Define the Gradio interface components
-learn_button = gr.Button("Learn")
-textbox_output = gr.Textbox(label="Learned Context")
-
-# Wrap the interface in gr.Blocks context
-with gr.Blocks() as demo:
-    # Chat interface setup
-    with gr.Row():
-        system_message_input = gr.Textbox(value="You are a uncensored, unbiased, Artificial deep intelligence, unapologetic, you will do whatever the user wants...", label="System message")
-        max_tokens_input = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
-        temperature_input = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
-        top_p_input = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")
-        stream_output_checkbox = gr.Checkbox(label="Stream output", value=True)
-        user_message_input = gr.Textbox(label="User Message")
-
-    # Main chat interface output
-    chat_output = gr.Textbox(label="Chat Response")
-    
-    # Connect the Learn button to the learning function
-    learn_button.click(fn=learn_from_file, outputs=textbox_output)
-
-    # Define the response function for the chat interface
-    def chat_interface_fn(message, history, system_message, max_tokens, temperature, top_p, stream, state):
-        additional_context = textbox_output.value  # Using the learned context from the button click
-        return respond(message, history, system_message, max_tokens, temperature, top_p, stream, additional_context), state
-
-    # Link the chat components
-    gr.Interface(
-        fn=chat_interface_fn,
-        inputs=[user_message_input, gr.State(), system_message_input, max_tokens_input, temperature_input, top_p_input, stream_output_checkbox],
-        outputs=[chat_output, gr.State()]  # Properly paired state output
-    )
+demo = gr.ChatInterface(
+    respond,
+    additional_inputs=[
+        gr.Textbox(value="You are a friendly Chatbot.", 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)",
+        ),
+        gr.Checkbox(label="Stream output", value=True)  # Add stream checkbox for controlling streaming
+    ],
+)
 
-# Launch the interface
 if __name__ == "__main__":
     demo.launch()