About this app
Testing and benchmarking LLM agents including behavioral testing, capability assessment, reliability metrics, and production monitoring—where even top agents achieve less than 50% on real-world benchm
🔌
需要 ClawBox 硬件
在您自己的 AI 硬件上本地运行 Agent Evaluation — 私密、快速、无需云端。
