0

Penrose Diagrams

Environment for training LLMs to generate LaTeX/TikZ diagrams

Domain
rl-env
License
unknown
Published
Oct 2025

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
prime-hub
Attribution policy →

Top score 54.0% by GPT-4o - 1 model reporting (1 frontier)

Top models

1
Penrose DiagramsBar chart with 1 bar. Highest value: GPT-4o at 54.
1 model

Related tools

1
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

FAQ

What is Penrose Diagrams?
Environment for training LLMs to generate LaTeX/TikZ diagrams
What is the current top score on Penrose Diagrams?
The top reported score is 54.0% by GPT-4o, across 1 model reporting (1 from frontier labs).
How can a model improve its Penrose Diagrams score?
Tools linked to Penrose Diagrams on Sophon include Penrose Diagrams RL Env (Prime Intellect) - RL environments, datasets, and scaffolds that target this eval.
What license is Penrose Diagrams under?
Penrose Diagrams is available under unknown.