Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| import os | |
| from huggingface_hub import InferenceClient | |
| from utils import validate_token, format_chat_history | |
| from config import DEFAULT_MODEL, DEFAULT_PROVIDER, MAX_TOKENS, TEMPERATURE | |
| # Initialize the client with secure token handling | |
| def get_client(): | |
| """Initialize and return the Hugging Face Inference Client""" | |
| token = os.getenv("HF_TOKEN") | |
| if not token: | |
| raise ValueError("HF_TOKEN environment variable is not set") | |
| return InferenceClient(token=token) | |
| def chat_response(message, history, model, provider, temperature, max_tokens): | |
| """ | |
| Generate chat response using Hugging Face Inference API | |
| """ | |
| try: | |
| client = get_client() | |
| # Format the history for the API | |
| messages = format_chat_history(history, message) | |
| # Make the API call - try different approaches for provider | |
| try: | |
| # First try with provider parameter (newer versions) | |
| response = client.chat_completion( | |
| model=model, | |
| messages=messages, | |
| max_tokens=max_tokens, | |
| temperature=temperature, | |
| stream=False | |
| ) | |
| except TypeError as e: | |
| if "provider" in str(e): | |
| # If provider parameter not supported, try without it | |
| # Some providers might be specified in the model name or URL | |
| response = client.chat_completion( | |
| model=f"{provider}/{model}" if provider else model, | |
| messages=messages, | |
| max_tokens=max_tokens, | |
| temperature=temperature, | |
| stream=False | |
| ) | |
| else: | |
| raise e | |
| # Extract the assistant's response | |
| assistant_message = response.choices[0].message.content | |
| return assistant_message | |
| except Exception as e: | |
| return f"Error: {str(e)}" | |
| def user_message(user_message, history): | |
| """Add user message to history and return empty string""" | |
| return "", history + [{"role": "user", "content": user_message}] | |
| def bot_message(history, model, provider, temperature, max_tokens): | |
| """Generate bot response and add to history""" | |
| try: | |
| if not history: | |
| return history | |
| # Get the last user message | |
| last_message = history[-1]["content"] if history else "" | |
| # Get previous messages (excluding the current user message) | |
| previous_history = history[:-1] if len(history) > 1 else [] | |
| # Generate response | |
| response = chat_response(last_message, previous_history, model, provider, temperature, max_tokens) | |
| # Add assistant response to history | |
| history.append({"role": "assistant", "content": response}) | |
| return history | |
| except Exception as e: | |
| history.append({"role": "assistant", "content": f"Error: {str(e)}"}) | |
| return history | |
| def create_interface(): | |
| """Create and configure the Gradio interface""" | |
| with gr.Blocks( | |
| title="Hugging Face Chat Interface", | |
| theme=gr.themes.Soft(), | |
| css=""" | |
| .gradio-container { | |
| max-width: 800px; | |
| margin: auto; | |
| } | |
| """ | |
| ) as demo: | |
| gr.Markdown(""" | |
| # 🤖 Hugging Face Chat Interface | |
| Built with [anycoder](https://huggingface.co/spaces/akhaliq/anycoder) | |
| Chat with Llama-3.1-8B-Instruct using Hugging Face Inference API. | |
| """) | |
| with gr.Row(): | |
| with gr.Column(scale=1): | |
| model_input = gr.Textbox( | |
| value=DEFAULT_MODEL, | |
| label="Model", | |
| placeholder="Enter model name" | |
| ) | |
| provider_input = gr.Textbox( | |
| value=DEFAULT_PROVIDER, | |
| label="Provider", | |
| placeholder="Enter provider name (may not be supported by all API versions)" | |
| ) | |
| with gr.Column(scale=1): | |
| temperature_slider = gr.Slider( | |
| minimum=0.0, | |
| maximum=2.0, | |
| value=TEMPERATURE, | |
| step=0.1, | |
| label="Temperature" | |
| ) | |
| max_tokens_slider = gr.Slider( | |
| minimum=1, | |
| maximum=4096, | |
| value=MAX_TOKENS, | |
| step=1, | |
| label="Max Tokens" | |
| ) | |
| chatbot = gr.Chatbot( | |
| type="messages", | |
| label="Chat", | |
| height=400, | |
| show_copy_button=True | |
| ) | |
| msg = gr.Textbox( | |
| label="Your message", | |
| placeholder="Type your message here...", | |
| lines=2 | |
| ) | |
| with gr.Row(): | |
| submit_btn = gr.Button("Send", variant="primary") | |
| clear_btn = gr.Button("Clear") | |
| # State to maintain chat history | |
| chat_history = gr.State([]) | |
| # Handle message submission - proper flow for type="messages" | |
| msg.submit( | |
| user_message, | |
| inputs=[msg, chat_history], | |
| outputs=[msg, chat_history] | |
| ).then( | |
| bot_message, | |
| inputs=[chat_history, model_input, provider_input, temperature_slider, max_tokens_slider], | |
| outputs=[chat_history] | |
| ).then( | |
| lambda history: history, | |
| inputs=[chat_history], | |
| outputs=[chatbot] | |
| ) | |
| submit_btn.click( | |
| user_message, | |
| inputs=[msg, chat_history], | |
| outputs=[msg, chat_history] | |
| ).then( | |
| bot_message, | |
| inputs=[chat_history, model_input, provider_input, temperature_slider, max_tokens_slider], | |
| outputs=[chat_history] | |
| ).then( | |
| lambda history: history, | |
| inputs=[chat_history], | |
| outputs=[chatbot] | |
| ) | |
| clear_btn.click( | |
| lambda: [], | |
| outputs=[chat_history] | |
| ).then( | |
| lambda: [], | |
| outputs=[chatbot] | |
| ) | |
| # Example inputs | |
| gr.Examples( | |
| examples=[ | |
| ["Hello! How are you?"], | |
| ["Can you explain quantum computing in simple terms?"], | |
| ["What are the benefits of renewable energy?"], | |
| ["Write a short poem about nature."] | |
| ], | |
| inputs=[msg] | |
| ) | |
| return demo | |
| if __name__ == "__main__": | |
| # Validate environment setup | |
| if not validate_token(): | |
| print("Warning: HF_TOKEN environment variable is not set.") | |
| print("Please set your Hugging Face token as an environment variable.") | |
| print("Example: export HF_TOKEN=your_token_here") | |
| demo = create_interface() | |
| demo.launch( | |
| share=False, | |
| show_error=True, | |
| show_api=True | |
| ) |