tabs/run_benchmark.py

import os
import importlib
import sys
# from pathlib import Path
from benchmark.run_benchmark import run_benchmark
# from dotenv import load_dotenv


# file_path = Path(__file__).resolve()
# parent_path = file_path.parent.parent
# load_dotenv(parent_path / ".env")


def run_benchmark_main(tool_name, model_name, openai_api_key, anthropic_api_key):
    """Run the benchmark using the provided function and API key."""
    # Empyt the results directory
    os.system("rm -rf results/*")

    # Set the benchmark parameters
    kwargs = {}
    kwargs["num_questions"] = 2
    kwargs["tools"] = [tool_name]
    if model_name:
        kwargs["model"] = [model_name]
    kwargs["api_keys"] = {}
    if openai_api_key:
        kwargs["api_keys"]["openai"] = openai_api_key
    if anthropic_api_key:
        kwargs["api_keys"]["anthropic"] = anthropic_api_key

    kwargs["num_urls"] = 3
    kwargs["num_words"] = 300
    kwargs["provide_source_links"] = True

    print(f"Running benchmark with the following parameters: {kwargs}")
    
    # Run the benchmark
    try:
        run_benchmark(kwargs=kwargs)
        return "completed"
    except Exception as e:
        return f"Error running benchmark: {e}"