Update app.py

app.py

@@ -3,11 +3,15 @@ import os
 import sys
 import json 
 import requests
+import random
+
 
 MODEL = "gpt-4o-mini"
 API_URL = os.getenv("API_URL")
 DISABLED = os.getenv("DISABLED") == 'True'
-OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
+OPENAI_API_KEYS = os.getenv("OPENAI_API_KEYS").split(',')
+print (API_URL)
+print (OPENAI_API_KEYS)
 NUM_THREADS = int(os.getenv("NUM_THREADS"))
 
 print (NUM_THREADS)
@@ -16,20 +20,6 @@ def exception_handler(exception_type, exception, traceback):
     print("%s: %s" % (exception_type.__name__, exception))
 sys.excepthook = exception_handler
 sys.tracebacklimit = 0
-
-#https://github.com/gradio-app/gradio/issues/3531#issuecomment-1484029099
-def parse_codeblock(text):
-    lines = text.split("\n")
-    for i, line in enumerate(lines):
-        if "```" in line:
-            if line != "```":
-                lines[i] = f'<pre><code class="{lines[i][3:]}">'
-            else:
-                lines[i] = '</code></pre>'
-        else:
-            if i > 0:
-                lines[i] = "<br/>" + line.replace("<", "&lt;").replace(">", "&gt;")
-    return "".join(lines)
     
 def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:gr.Request):
     payload = {
@@ -42,11 +32,13 @@ def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:
         "presence_penalty":0,
         "frequency_penalty":0,
     }
-
+    OPENAI_API_KEY = random.choice(OPENAI_API_KEYS)
+    print (OPENAI_API_KEY)
+    headers_dict = {key.decode('utf-8'): value.decode('utf-8') for key, value in request.headers.raw}
     headers = {
         "Content-Type": "application/json",
         "Authorization": f"Bearer {OPENAI_API_KEY}",
-        "Headers": f"{request.kwargs['headers']}"
+        "Headers": f"{headers_dict}"
     }
 
     # print(f"chat_counter - {chat_counter}")
@@ -93,6 +85,8 @@ def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:
         #    raise Exception(f"Sorry, hitting rate limit. Please try again later. {response}")
         
         for chunk in response.iter_lines():
+            print (chunk)
+            sys.stdout.flush()
             #Skipping first chunk
             if counter == 0:
                 counter += 1
@@ -109,19 +103,19 @@ def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:
                     else:
                         history[-1] = partial_words
                     token_counter += 1
-                    yield [(parse_codeblock(history[i]), parse_codeblock(history[i + 1])) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=False), gr.update(interactive=False)  # resembles {chatbot: chat, state: history}  
+                    yield [(history[i], history[i + 1]) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=False), gr.update(interactive=False)  # resembles {chatbot: chat, state: history}  
     except Exception as e:
         print (f'error found: {e}')
-    yield [(parse_codeblock(history[i]), parse_codeblock(history[i + 1])) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=True), gr.update(interactive=True)
+    yield [(history[i], history[i + 1]) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=True), gr.update(interactive=True)
     print(json.dumps({"chat_counter": chat_counter, "payload": payload, "partial_words": partial_words, "token_counter": token_counter, "counter": counter}))
                    
 
 def reset_textbox():
     return gr.update(value='', interactive=False), gr.update(interactive=False)
 
-title = """<h1 align="center">GPT-4o mini Chatbot</h1>"""
+title = """<h1 align="center">GPT-4o: Research Preview (Short-Term Availability)</h1>"""
 if DISABLED:
-    title = """<h1 align="center" style="color:red">This app has reached OpenAI's usage limit. We are currently requesting an increase in our quota. Please check back in a few days.</h1>"""
+    title = """<h1 align="center" style="color:red">This app has reached OpenAI's usage limit. Please check back tomorrow.</h1>"""
 description = """Language models can be conditioned to act like dialogue agents through a conversational prompt that typically takes the form:
 ```
 User: <utterance>
@@ -130,7 +124,7 @@ User: <utterance>
 Assistant: <utterance>
 ...
 ```
-In this app, you can explore the outputs of a gpt-4o mini LLM.
+In this app, you can explore the outputs of a gpt-4 turbo LLM.
 """
 
 theme = gr.themes.Default(primary_hue="green")                
@@ -139,17 +133,18 @@ with gr.Blocks(css = """#col_container { margin-left: auto; margin-right: auto;}
                 #chatbot {height: 520px; overflow: auto;}""",
               theme=theme) as demo:
     gr.HTML(title)
-    gr.HTML("""<h3 align="center">This app provides you full access to GPT-4o mini (128K token limit). You don't need any OPENAI API key.</h1>""")
-    #gr.HTML('''<center><a href="https://huggingface.co/spaces/yuntian-deng/ChatGPT?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>Duplicate the Space and run securely with your OpenAI API Key</center>''')
+    gr.HTML("""<h3 align="center">This app provides you full access to GPT-4o mini (128K token limit). You don't need any OPENAI API key.</h3>""")
+
+    #gr.HTML('''<center><a href="https://huggingface.co/spaces/ysharma/ChatGPT4?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>Duplicate the Space and run securely with your OpenAI API Key</center>''')
     with gr.Column(elem_id = "col_container", visible=False) as main_block:
-        #API Key is provided by OpenAI 
-        #openai_api_key = gr.Textbox(type='password', label="Enter only your OpenAI API key here")
+        #GPT4 API Key is provided by Huggingface 
+        #openai_api_key = gr.Textbox(type='password', label="Enter only your GPT4 OpenAI API key here")
         chatbot = gr.Chatbot(elem_id='chatbot') #c
         inputs = gr.Textbox(placeholder= "Hi there!", label= "Type an input and press Enter") #t
         state = gr.State([]) #s
         with gr.Row():
             with gr.Column(scale=7):
-                b1 = gr.Button(visible=not DISABLED).style(full_width=True)
+                b1 = gr.Button(visible=not DISABLED) #.style(full_width=True)
             with gr.Column(scale=3):
                 server_status_code = gr.Textbox(label="Status code from OpenAI server", )
     
@@ -181,9 +176,9 @@ with gr.Blocks(css = """#col_container { margin-left: auto; margin-right: auto;}
             accept_button = gr.Button("I Agree")
 
         def enable_inputs():
-            return user_consent_block.update(visible=False), main_block.update(visible=True)
+            return gr.update(visible=False), gr.update(visible=True)
 
-    accept_button.click(None, None, accept_checkbox, _js=js, queue=False)
+    accept_button.click(None, None, accept_checkbox, js=js, queue=False)
     accept_checkbox.change(fn=enable_inputs, inputs=[], outputs=[user_consent_block, main_block], queue=False)
 
     inputs.submit(reset_textbox, [], [inputs, b1], queue=False)
@@ -191,4 +186,4 @@ with gr.Blocks(css = """#col_container { margin-left: auto; margin-right: auto;}
     b1.click(reset_textbox, [], [inputs, b1], queue=False)
     b1.click(predict, [inputs, top_p, temperature, chat_counter, chatbot, state], [chatbot, state, chat_counter, server_status_code, inputs, b1],)  #openai_api_key
              
-    demo.queue(max_size=20, concurrency_count=NUM_THREADS, api_open=False).launch(share=False)
+    demo.queue(max_size=10, default_concurrency_limit=NUM_THREADS, api_open=False).launch(share=False)