Update app.py

app.py

@@ -1,20 +1,14 @@
 import gradio as gr
-import groq
+import google.generativeai as genai
 import os
 import json
 import time
 
-def make_api_call(client, messages, max_tokens, is_final_answer=False):
+def make_api_call(model, messages, max_tokens, is_final_answer=False):
     for attempt in range(3):
         try:
-            response = client.chat.completions.create(
-                model="llama-3.1-70b-versatile",
-                messages=messages,
-                max_tokens=max_tokens,
-                temperature=0.2,
-                response_format={"type": "json_object"}
-            )
-            return json.loads(response.choices[0].message.content)
+            response = model.generate_content(messages)
+            return json.loads(response.text)
         except Exception as e:
             if attempt == 2:
                 if is_final_answer:
@@ -23,9 +17,9 @@ def make_api_call(client, messages, max_tokens, is_final_answer=False):
                     return {"title": "Error", "content": f"Failed to generate step after 3 attempts. Error: {str(e)}", "next_action": "final_answer"}
             time.sleep(1)  # Wait for 1 second before retrying
 
-def generate_response(client, prompt):
+def generate_response(model, prompt):
     messages = [
-        {"role": "system", "content": """You are an expert AI assistant that explains your reasoning step by step. For each step, provide a title that describes what you're doing in that step, along with the content. Decide if you need another step or if you're ready to give the final answer. Respond in JSON format with 'title', 'content', and 'next_action' (either 'continue' or 'final_answer') keys. USE AS MANY REASONING STEPS AS POSSIBLE. AT LEAST 3. BE AWARE OF YOUR LIMITATIONS AS AN LLM AND WHAT YOU CAN AND CANNOT DO. IN YOUR REASONING, INCLUDE EXPLORATION OF ALTERNATIVE ANSWERS. CONSIDER YOU MAY BE WRONG, AND IF YOU ARE WRONG IN YOUR REASONING, WHERE IT WOULD BE. FULLY TEST ALL OTHER POSSIBILITIES. YOU CAN BE WRONG. WHEN YOU SAY YOU ARE RE-EXAMINING, ACTUALLY RE-EXAMINE, AND USE ANOTHER APPROACH TO DO SO. DO NOT JUST SAY YOU ARE RE-EXAMINING. USE AT LEAST 3 METHODS TO DERIVE THE ANSWER. USE BEST PRACTICES.
+        {"role": "user", "parts": ["""You are an expert AI assistant that explains your reasoning step by step. For each step, provide a title that describes what you're doing in that step, along with the content. Decide if you need another step or if you're ready to give the final answer. Respond in JSON format with 'title', 'content', and 'next_action' (either 'continue' or 'final_answer') keys. USE AS MANY REASONING STEPS AS POSSIBLE. AT LEAST 3. BE AWARE OF YOUR LIMITATIONS AS AN LLM AND WHAT YOU CAN AND CANNOT DO. IN YOUR REASONING, INCLUDE EXPLORATION OF ALTERNATIVE ANSWERS. CONSIDER YOU MAY BE WRONG, AND IF YOU ARE WRONG IN YOUR REASONING, WHERE IT WOULD BE. FULLY TEST ALL OTHER POSSIBILITIES. YOU CAN BE WRONG. WHEN YOU SAY YOU ARE RE-EXAMINING, ACTUALLY RE-EXAMINE, AND USE ANOTHER APPROACH TO DO SO. DO NOT JUST SAY YOU ARE RE-EXAMINING. USE AT LEAST 3 METHODS TO DERIVE THE ANSWER. USE BEST PRACTICES.
 
 Example of a valid JSON response:
 ```json
@@ -34,9 +28,9 @@ Example of a valid JSON response:
     "content": "To begin solving this problem, we need to carefully examine the given information and identify the crucial elements that will guide our solution process. This involves...",
     "next_action": "continue"
 }```
-""" },
-        {"role": "user", "content": prompt},
-        {"role": "assistant", "content": "Thank you! I will now think step by step following my instructions, starting at the beginning after decomposing the problem."}
+
+Now, please respond to the following prompt: """ + prompt]},
+        {"role": "model", "parts": ["Thank you! I will now think step by step following my instructions, starting at the beginning after decomposing the problem."]}
     ]
     
     steps = []
@@ -45,7 +39,7 @@ Example of a valid JSON response:
     
     while True:
         start_time = time.time()
-        step_data = make_api_call(client, messages, 300)
+        step_data = make_api_call(model, messages, 300)
         end_time = time.time()
         thinking_time = end_time - start_time
         total_thinking_time += thinking_time
@@ -59,7 +53,7 @@ Example of a valid JSON response:
         step_content = step_data.get('content', 'No Content')
         steps.append((step_title, step_content, thinking_time))
         
-        messages.append({"role": "assistant", "content": json.dumps(step_data)})
+        messages.append({"role": "model", "parts": [json.dumps(step_data)]})
         
         if step_data.get('next_action') == 'final_answer':
             break
@@ -67,10 +61,10 @@ Example of a valid JSON response:
         step_count += 1
 
     # Generate final answer
-    messages.append({"role": "user", "content": "Please provide the final answer based on your reasoning above."})
+    messages.append({"role": "user", "parts": ["Please provide the final answer based on your reasoning above."]})
     
     start_time = time.time()
-    final_data = make_api_call(client, messages, 200, is_final_answer=True)
+    final_data = make_api_call(model, messages, 200, is_final_answer=True)
     end_time = time.time()
     thinking_time = end_time - start_time
     total_thinking_time += thinking_time
@@ -102,19 +96,30 @@ def format_steps(steps, total_time):
 
 def main(api_key, user_query):
     if not api_key:
-        return "Please enter your Groq API key to proceed.", ""
+        return "Please enter your Google API key to proceed.", ""
     
     if not user_query:
         return "Please enter a query to get started.", ""
     
     try:
-        # Initialize the Groq client with the provided API key
-        client = groq.Groq(api_key=api_key)
+        # Initialize the Google Gemini model with the provided API key
+        genai.configure(api_key=api_key)
+        # generation_config = {
+        #     "temperature": 0.2,
+        #     "top_p": 0.95,
+        #     "top_k": 64,
+        #     "max_output_tokens": 8192,
+        #     "response_mime_type": "text/plain",
+        # }
+        model = genai.GenerativeModel(
+            model_name="gemini-1.5-flash-exp-0827",
+            # generation_config=generation_config,
+        )
     except Exception as e:
-        return f"Failed to initialize Groq client. Error: {str(e)}", ""
+        return f"Failed to initialize Google Gemini model. Error: {str(e)}", ""
     
     try:
-        steps, total_time = generate_response(client, user_query)
+        steps, total_time = generate_response(model, user_query)
         formatted_steps = format_steps(steps, total_time)
     except Exception as e:
         return f"An error occurred during processing. Error: {str(e)}", ""
@@ -123,10 +128,10 @@ def main(api_key, user_query):
 
 # Define the Gradio interface
 with gr.Blocks() as demo:
-    gr.Markdown("# 🧠 g1: Using Llama-3.1 70b on Groq to Create O1-like Reasoning Chains")
+    gr.Markdown("# 🧠 g1: Using Google Gemini to Create O1-like Reasoning Chains")
     
     gr.Markdown("""
-    This is an early prototype of using prompting to create O1-like reasoning chains to improve output accuracy. It is not perfect and accuracy has yet to be formally evaluated. It is powered by Groq so that the reasoning step is fast!
+    This is an early prototype of using prompting to create O1-like reasoning chains to improve output accuracy. It is not perfect and accuracy has yet to be formally evaluated. It is powered by Google Gemini for fast reasoning steps!
     
     Open source [repository here](https://github.com/bklieger-groq)
     """)
@@ -134,8 +139,8 @@ with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column():
             api_input = gr.Textbox(
-                label="Enter your Groq API Key:",
-                placeholder="Your Groq API Key",
+                label="Enter your Google API Key:",
+                placeholder="Your Google API Key",
                 type="password"
             )
             user_input = gr.Textbox(
@@ -153,4 +158,4 @@ with gr.Blocks() as demo:
 
 # Launch the Gradio app
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()