uhiguys/cad-bench-ed-2026-anonymous

该数据集包含CAD-bench的公共任务载荷，CAD-bench是一个用于语言模型CAD代理的可执行基准。每个任务目录包括：自然语言基准提示（prompt.txt）、任务元数据（task.toml）、用于验证和媒体生成的参考Build123D解决方案（gold.py），以及可选的STEP文件或Blender模拟脚本等固定装置。数据集旨在与CAD-bench运行时一起使用，以评估CAD代码生成或代理CAD系统。数据集包含17个任务，其中一些简单的几何任务已接近被当前模型解决，而功能性装配任务仍然具有挑战性。当前参考实现使用Build123D，但基准旨在评分提交的CAD工件，而不是特定的建模API。

This dataset contains the public task payloads for CAD-bench, an executable benchmark for language-model CAD agents. Each task directory includes: the natural-language benchmark prompt (prompt.txt), task metadata (task.toml), a reference Build123D solution used for validation and media generation (gold.py), and optional fixtures such as STEP files or Blender simulation scripts. The dataset is intended to be used with the CAD-bench runtime to evaluate CAD code-generation or agentic CAD systems. The release has 17 tasks, with some simple geometry tasks close to being solved by current models, while functional assembly tasks remain difficult. The current reference implementations use Build123D, although the benchmark is intended to score submitted CAD artifacts rather than a particular modeling API.