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/pawbenchDisplaying 0 of 150 tasks
PawBench: A comprehensive agent benchmark with 150 tasks covering coding, data analysis, tool use, reasoning, safety, and multimodal capabilities.
harbor run -d agentscope-ai/pawbenchDisplaying 0 of 150 tasks
PawBench: A comprehensive agent benchmark with 150 tasks covering coding, data analysis, tool use, reasoning, safety, and multimodal capabilities.
harbor run -d agentscope-ai/pawbenchDisplaying 0 of 150 tasks