Buckets:
910 MB
22 files
Updated 6 days ago
Ctrl+K
| Name | Size | Uploaded | Xet hash |
|---|---|---|---|
| QA | 1 items | ||
| code_gen | 18 items | ||
| edit-bench | 1 items | ||
| .gitattributes | 2.5 kB xet | 738f1125 | |
| README.md | 1.8 kB xet | b95749ca |
BenchCAD
Three-config dataset for CAD evaluation:
- edit-bench — held-out CAD edit benchmark.
- code_gen — 17,900 synthetic CadQuery samples (compact 12-column variant) covering 106 mechanical part families. Each row contains the GT CadQuery code plus 5 normalized renders.
- QA — CAD question-answering benchmark.
code_gen schema (12 columns)
| Column | Type | Description |
|---|---|---|
stem |
string | unique sample identifier |
family |
string | mechanical part family (106 distinct) |
variant |
string | sub-variant within family |
difficulty |
string | easy / medium / hard |
base_plane |
string | initial workplane (XY / XZ / YZ) |
standard |
string | ISO/DIN standard if applicable |
code |
string | CadQuery Python source (ground truth) |
view_0_png |
image | front view (134×134 PNG) |
view_1_png |
image | right view |
view_2_png |
image | top view |
view_3_png |
image | iso view |
composite_png |
image | 2×2 composite of the four views |
Usage
from datasets import load_dataset
ds = load_dataset("BenchCAD/BenchCAD", "code_gen", split="code_gen")
print(ds[0]["family"], ds[0]["difficulty"])
ds[0]["composite_png"].show()
print(ds[0]["code"])
- Total size
- 910 MB
- Files
- 22
- Last updated
- Jun 23
- Pre-warmed CDN
- US EU US EU