agentscope-ai/pawbench

harbor run -d agentscope-ai/pawbench

PawBench: A comprehensive agent benchmark with 150 tasks covering coding, data analysis, tool use, reasoning, safety, and multimodal capabilities.

harbor run -d agentscope-ai/pawbench

⌘K

PawBench: A comprehensive agent benchmark with 150 tasks covering coding, data analysis, tool use, reasoning, safety, and multimodal capabilities.

harbor run -d agentscope-ai/pawbench

⌘K

agentscope-ai/pawbench | Harbor