LLM BASELINEGoogle · 2025

Gemini 2.0 (zero-shot)

Gemini 2.0 prompted zero-shot on the CADPrompt benchmark achieves 85% compile rate. Included as a second general-purpose LLM baseline alongside GPT-4o and Claude.

Input
Text
Output
CadQuery
Venue
Google
Year
2025
CAD Arena Status
Scheduled for evaluation on the full 200-prompt benchmark. Results will appear on the leaderboard at launch.