app.py

import gradio as gr
from huggingface_hub import InferenceClient

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""

# Default client with the first model
client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.3")

# Function to switch between models based on selection
def switch_client(model_name: str):
    return InferenceClient(model_name)

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    model_name  # Add this parameter for model selection
):
    # Switch client based on model selection
    global client
    client = switch_client(model_name)
    
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    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
        yield response

# Model names and their pseudonyms
model_choices = [
    ("mistralai/Mistral-7B-Instruct-v0.3", "Lake 1 Base")
]

# Convert pseudonyms to model names for the dropdown
pseudonyms = [model[1] for model in model_choices]

# Function to handle model selection and pseudonyms
def respond_with_pseudonym(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    selected_pseudonym
):
    # Find the actual model name from the pseudonym
    model_name = next(model[0] for model in model_choices if model[1] == selected_pseudonym)
    
    # Call the existing respond function
    response = list(respond(message, history, system_message, max_tokens, temperature, top_p, model_name))
    
    # Add pseudonym at the end of the response
    response[-1] += f"\n\n[Response generated by: {selected_pseudonym}]"
    
    return response

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond_with_pseudonym,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
        gr.Dropdown(pseudonyms, label="Select Model", value=pseudonyms[0])  # Pseudonym selection dropdown
    ],
)

if __name__ == "__main__":
    demo.launch()