Update app.py

app.py

@@ -1,15 +1,19 @@
 import gradio as gr
-import requests
+from openai import OpenAI
 import os
 
 ACCESS_TOKEN = os.getenv("HF_TOKEN")
+print("Access token loaded.")
 
-API_URL = "https://api-inference.huggingface.co/v1/chat/completions"
-HEADERS = {"Authorization": f"Bearer {ACCESS_TOKEN}"}
+client = OpenAI(
+    base_url="https://api-inference.huggingface.co/v1/",
+    api_key=ACCESS_TOKEN,
+)
+print("OpenAI client initialized.")
 
 def respond(
     message,
-    history,
+    history: list[tuple[str, str]],
     system_message,
     max_tokens,
     temperature,
@@ -18,67 +22,69 @@ def respond(
     seed,
     custom_model
 ):
+    print(f"Received message: {message}")
+    print(f"History: {history}")
+    print(f"System message: {system_message}")
+    
     if seed == -1:
         seed = None
     
     messages = [{"role": "system", "content": system_message}]
+    
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
         if val[1]:
             messages.append({"role": "assistant", "content": val[1]})
-    messages.append({"role": "user", "content": message})
     
-    model_to_use = custom_model.strip() if custom_model.strip() else "meta-llama/Llama-3.2-3B-Instruct"
+    messages.append({"role": "user", "content": message})
     
-    payload = {
-        "model": model_to_use,
-        "messages": messages,
-        "max_tokens": max_tokens,
-        "temperature": temperature,
-        "top_p": top_p,
-        "frequency_penalty": frequency_penalty,
-        "seed": seed,
-    }
+    model_to_use = custom_model.strip() if custom_model.strip() != "" else "meta-llama/Llama-3.2-3B-Instruct"
     
-    response = requests.post(API_URL, headers=HEADERS, json=payload)
+    response = ""
     
-    if response.status_code == 200:
-        return response.json()["choices"][0]["message"]["content"]
-    else:
-        return "Error: " + response.text
+    for message_chunk in client.chat.completions.create(
+        model=model_to_use,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+        frequency_penalty=frequency_penalty,
+        seed=seed,
+        messages=messages,
+    ):
+        token_text = message_chunk.choices[0].delta.content
+        response += token_text
+        yield response
 
 chatbot = gr.Chatbot(height=600, show_copy_button=True, placeholder="ChatGPT is initializing...", likeable=True, layout="panel")
 
-system_message_box = gr.Textbox(value="System message here", label="System Message")
-max_tokens_slider = gr.Slider(1024, 2048, value=1024, step=100, label="Max Tokens")
+system_message_box = gr.Label(value="You can select Max Tokens, Temperature, Top-P, Seed")
+
+max_tokens_slider = gr.Slider(1024, 2048, value=1024, step=100, label="Max new tokens")
 temperature_slider = gr.Slider(0.1, 1.0, value=0.7, step=0.1, label="Temperature")
 top_p_slider = gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-P")
 frequency_penalty_slider = gr.Slider(-2.0, 2.0, value=0.0, step=0.1, label="Frequency Penalty")
 seed_slider = gr.Slider(-1, 65535, value=-1, step=1, label="Seed (-1 for random)")
-custom_model_box = gr.Textbox(value="meta-llama/Llama-3.2-3B-Instruct", label="Custom Model")
 
-voice_input = gr.Microphone(label="Speak your message")
+custom_model_box = gr.Textbox(value="meta-llama/Llama-3.2-3B-Instruct", label="AI Mode is ")
 
-with gr.Blocks() as demo:
-    gr.Markdown("# AI Chatbot with Voice Input")
-    chatbot.render()
+demo = gr.ChatInterface(
+    fn=respond,
+    additional_inputs=[
+        system_message_box,
+        max_tokens_slider,
+        temperature_slider,
+        top_p_slider,
+        frequency_penalty_slider,
+        seed_slider,
+        custom_model_box,
+    ],
+    fill_height=True,
+    chatbot=chatbot,
     
-    with gr.Row():
-        system_message_box.render()
-        max_tokens_slider.render()
-        temperature_slider.render()
-        top_p_slider.render()
-    
-    with gr.Row():
-        frequency_penalty_slider.render()
-        seed_slider.render()
-        custom_model_box.render()
-    
-    voice_input.render()
-    
-    voice_input.change(respond, inputs=[voice_input, chatbot, system_message_box, max_tokens_slider, temperature_slider, top_p_slider, frequency_penalty_slider, seed_slider, custom_model_box], outputs=chatbot)
+)
 
 if __name__ == "__main__":
+    print("Launching the ChatGPT-Llama...")
     demo.launch()
-