Quickstart

Installation

pip install benchflow

Get your benchflow key at Benchflow.ai

Test your agent

from benchflow import load_benchmark

bench = load_benchmark(benchmark_name="webarena", bf_token="your_benchflow_key")

your_agents = youragent()

job_ids = bench.run(
    agents=your_agents,
    api={"provider": "openai", "model": "gpt-4o-mini", "OPENAI_API_KEY": os.getenv("OPENAI_API_KEY")},
    requirements_txt="webarena_requirements.txt",
    params={}
)

results = bench.get_results(job_ids)
python test.py

Last updated