Spaces:
Running
Running
| import os | |
| import gradio as gr | |
| from openai import OpenAI | |
| from typing import List, Tuple | |
| # Define available models | |
| AVAILABLE_MODELS = { | |
| "Sonar Pro": "sonar-pro", | |
| "Sonar": "sonar", | |
| } | |
| PX_ENDPOINT_URL = "https://api.perplexity.ai" | |
| PX_API_KEY = os.getenv('PX_KEY') | |
| PASSWORD = os.getenv("PASSWD") # Store the password in an environment variable | |
| px_client = OpenAI(base_url=PX_ENDPOINT_URL, api_key=PX_API_KEY) | |
| def respond( | |
| message: str, | |
| history: List[Tuple[str, str]], | |
| system_message: str, | |
| model_choice: str, | |
| max_tokens: int, | |
| temperature: float, | |
| top_p: float, | |
| ): | |
| messages = [{"role": "system", "content": system_message}] | |
| for user_msg, assistant_msg in history: | |
| if user_msg: | |
| messages.append({"role": "user", "content": user_msg}) | |
| if assistant_msg: | |
| messages.append({"role": "assistant", "content": assistant_msg}) | |
| messages.append({"role": "user", "content": message}) | |
| response = "" | |
| citations = [] | |
| stream = px_client.chat.completions.create( | |
| model=AVAILABLE_MODELS[model_choice], | |
| messages=messages, | |
| max_tokens=max_tokens, | |
| temperature=temperature, | |
| top_p=top_p, | |
| stream=True, | |
| ) | |
| for chunk in stream: | |
| if "choices" in chunk: | |
| token = chunk.choices[0].delta.content or "" | |
| response += token | |
| yield response # Stream response as it arrives | |
| if "citations" in chunk: | |
| citations = chunk["citations"] | |
| # Append citations as clickable links | |
| if citations: | |
| citation_text = "\n\nSources:\n" + "\n".join( | |
| [f"[{i+1}] [{url}]({url})" for i, url in enumerate(citations)] | |
| ) | |
| response += citation_text | |
| yield response | |
| def check_password(input_password): | |
| if input_password == PASSWORD: | |
| return gr.update(visible=False), gr.update(visible=True) | |
| else: | |
| return gr.update(value="", interactive=True), gr.update(visible=False) | |
| with gr.Blocks() as demo: | |
| with gr.Column(): | |
| password_input = gr.Textbox( | |
| type="password", label="Enter Password", interactive=True | |
| ) | |
| submit_button = gr.Button("Submit") | |
| error_message = gr.Textbox( | |
| label="Error", visible=False, interactive=False | |
| ) | |
| with gr.Column(visible=False) as chat_interface: | |
| system_prompt = gr.Textbox( | |
| value="You are a helpful assistant.", label="System message" | |
| ) | |
| chat = gr.ChatInterface( | |
| respond, | |
| chatbot=gr.Chatbot(height=400), # Set the desired height here | |
| additional_inputs=[] | |
| ) | |
| with gr.Column(): | |
| model_choice = gr.Dropdown( | |
| choices=list(AVAILABLE_MODELS.keys()), | |
| value=list(AVAILABLE_MODELS.keys())[0], | |
| label="Select Model" | |
| ) | |
| max_tokens = gr.Slider( | |
| minimum=1, maximum=30000, value=2048, step=100, 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 (nucleus sampling)" | |
| ) | |
| submit_button.click( | |
| check_password, inputs=password_input, outputs=[password_input, chat_interface] | |
| ) | |
| if __name__ == "__main__": | |
| demo.launch(share=True) | |