BREAKING
Sonnet 5 Trails GLM-5.2 in 3D Test
The Monica's apt test
1Read floor plan
2Generate threejs code
3Door placement off
4Chair orientation off
Open weight vs proprietary
GLM-5.2Open-weight
MIT license
Self-host and fine-tune
From Zhipu AI
Claude Sonnet 5Proprietary
Anthropic
No self-hosting
Closed weights
Output cost per 1M tokens
GLM-5.2 out6
Sonnet 5 out10
0%
GLM-5.2 F1
0%
Claude Code F1
Open models close the gap
AI NEWS BLITZ
Claude Sonnet 5 was judged worse than the open-weight GLM-5.2 in a 3D recreation test.