Comment by vunderba

There's also an interesting set of benchmarks around giving LLMs a sort of "building-block harness and tools" to see how well they do putting together voxel-style models.

https://github.com/Ammaar-Alam/minebench

These kinds of tests which build models programmatically (SVG construction, Blender-MCP, etc) also make it significantly easier to make or request adjustments to the final output.