app.py

import os
import gradio as gr

from openai import OpenAI

from optillm.cot_reflection import cot_reflection
from optillm.rto import round_trip_optimization
from optillm.z3_solver import Z3SolverSystem
from optillm.self_consistency import advanced_self_consistency_approach
from optillm.rstar import RStar
from optillm.plansearch import plansearch
from optillm.leap import leap


API_KEY = os.environ.get("OPENROUTER_API_KEY")

def respond(
    message,
    history: list[tuple[str, str]],
    model,
    approach,
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    client = OpenAI(api_key=API_KEY, base_url="https://openrouter.ai/api/v1")
    
    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})

    if approach == 'rto':
        final_response = round_trip_optimization(system_prompt, initial_query, client, model)
    elif approach == 'z3':
        z3_solver = Z3SolverSystem(system_prompt, client, model)
        final_response = z3_solver.process_query(initial_query)
    elif approach == "self_consistency":
        final_response = advanced_self_consistency_approach(system_prompt, initial_query, client, model)
    elif approach == "rstar":
        rstar = RStar(system_prompt, client, model)
        final_response = rstar.solve(initial_query)
    elif approach == "cot_reflection":
        final_response = cot_reflection(system_prompt, initial_query, client, model)
    elif approach == 'plansearch':
        final_response = plansearch(system_prompt, initial_query, client, model)
    elif approach == 'leap':
        final_response = leap(system_prompt, initial_query, client, model)
        
    return final_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

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Dropdown(
            ["nousresearch/hermes-3-llama-3.1-405b:free", "meta-llama/llama-3.1-8b-instruct:free", "qwen/qwen-2-7b-instruct:free",
            "google/gemma-2-9b-it:free", "mistralai/mistral-7b-instruct:free", ], 
            value="nousresearch/hermes-3-llama-3.1-405b:free", label="Model", info="Choose the base model"
        ),
        gr.Dropdown(
            ["leap", "plansearch", "rstar", "cot_reflection", "rto", "self_consistency", "z3"], value="cot_reflection", label="Approach", info="Choose the approach"
        ),
        gr.Textbox(value="", 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)",
        ),
    ],
)

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