Function handle_run

cli/main.py:409–435

(
    benchmark_file: str,
    benchmark_name: str,
    model_identifier: str | None,
    ip_addr: str,
    prompt_bundle_name: str = "default",
    config_file: str | None = None,
    output_file: str | None = None,
)

def handle_run(
    benchmark_file: str,
    benchmark_name: str,
    model_identifier: str | None,
    ip_addr: str,
    prompt_bundle_name: str = "default",
    config_file: str | None = None,
    output_file: str | None = None,
) -> int:
    config = _resolve_config(config_file)
    benchmark = find_benchmark_by_name(Path(benchmark_file), benchmark_name)
    model = _resolve_model_spec(model_identifier, config)
    request = build_task_request(
        benchmark=benchmark,
        model=model,
        ip_addr=ip_addr,
        config=config,
        prompt_bundle_name=prompt_bundle_name,
    )
    result = TaskRunner().run(request)
    payload = result.to_dict()

    if output_file:
        write_jsonl_record(output_file, payload)

    print(json.dumps(payload, ensure_ascii=False, indent=2))
    return 0 if result.status != "error" else 1


def handle_experiment(