它是否具备足够的智能体能力？在你自己的工具链上对开放模型进行基准测试

Is it agentic enough? Benchmarking open models on your own tooling