Spaces:
Sleeping
Sleeping
import gradio as gr | |
from huggingface_hub import InferenceClient | |
import time | |
# Initialize the client | |
client = InferenceClient("HuggingFaceH4/starchat2-15b-v0.1") | |
def respond( | |
message, | |
chat_history, | |
system_message, | |
max_tokens, | |
temperature, | |
top_p, | |
model_name | |
): | |
""" | |
Generate chat responses using the specified model. | |
""" | |
# Update client if model changes | |
global client | |
client = InferenceClient(model_name) | |
messages = [{"role": "system", "content": system_message}] | |
# Build conversation history | |
for human_msg, assistant_msg in chat_history: | |
if human_msg: | |
messages.append({"role": "user", "content": human_msg}) | |
if assistant_msg: | |
messages.append({"role": "assistant", "content": assistant_msg}) | |
messages.append({"role": "user", "content": message}) | |
response = "" | |
try: | |
for message in client.chat_completion( | |
messages, | |
max_tokens=max_tokens, | |
stream=True, | |
temperature=temperature, | |
top_p=top_p, | |
): | |
token = message.choices[0].delta.content | |
response += token | |
chat_history = chat_history + [(message, response)] | |
yield chat_history | |
except Exception as e: | |
chat_history = chat_history + [(message, f"Error: {str(e)}")] | |
yield chat_history | |
def create_chat_interface(): | |
""" | |
Create and configure the Gradio interface | |
""" | |
# Default system message | |
default_system = """You are a helpful AI assistant. You provide accurate, informative, and engaging responses while being direct and concise.""" | |
# Available models | |
models = [ | |
"HuggingFaceH4/starchat2-15b-v0.1", | |
"meta-llama/Llama-2-70b-chat-hf", | |
"mistralai/Mixtral-8x7B-Instruct-v0.1" | |
] | |
# Create the interface | |
with gr.Blocks(theme=gr.themes.Soft()) as demo: | |
gr.Markdown("# 🤖 Advanced AI Chatbot") | |
with gr.Row(): | |
with gr.Column(scale=3): | |
chatbot = gr.Chatbot( | |
height=600, | |
show_label=False, | |
container=True, | |
scale=2, | |
type="messages" # Set type to messages format | |
) | |
msg = gr.Textbox( | |
show_label=False, | |
placeholder="Type your message here...", | |
container=False | |
) | |
with gr.Column(scale=1): | |
with gr.Accordion("Settings", open=False): | |
system_msg = gr.Textbox( | |
label="System Message", | |
value=default_system, | |
lines=3 | |
) | |
model = gr.Dropdown( | |
choices=models, | |
value=models[0], | |
label="Model" | |
) | |
max_tokens = gr.Slider( | |
minimum=50, | |
maximum=4096, | |
value=1024, | |
step=1, | |
label="Max Tokens" | |
) | |
temperature = gr.Slider( | |
minimum=0.1, | |
maximum=2.0, | |
value=0.7, | |
step=0.1, | |
label="Temperature" | |
) | |
top_p = gr.Slider( | |
minimum=0.1, | |
maximum=1.0, | |
value=0.9, | |
step=0.1, | |
label="Top P" | |
) | |
with gr.Row(): | |
clear = gr.Button("Clear Chat") | |
stop = gr.Button("Stop") | |
# Initialize chat history | |
state = gr.State([]) | |
# Handle sending messages | |
msg.submit( | |
respond, | |
[msg, state, system_msg, max_tokens, temperature, top_p, model], | |
[state], | |
api_name="chat" | |
).then( | |
lambda x: "", | |
[msg], | |
[msg] | |
) | |
# Clear chat history | |
clear.click(lambda: [], None, state, queue=False) | |
# Example prompts | |
gr.Examples( | |
examples=[ | |
["Tell me a short story about a robot learning to paint."], | |
["Explain quantum computing in simple terms."], | |
["Write a haiku about artificial intelligence."] | |
], | |
inputs=msg | |
) | |
print(demo) | |
return demo | |
# Create and launch the interface | |
if __name__ == "__main__": | |
demo = create_chat_interface() | |
demo.queue() | |
# Disable SSR and sharing for Spaces | |
demo.launch( | |
share=False, # Disable sharing on Spaces | |
) | |