Phybench
PHYBench eval environment
- Domain
- rl-env
- License
- unknown
- Published
- Sep 2025
Cite
Notes
Only stored in your browser.
Top score 30.0% by GPT-4.1 Mini - 1 model reporting (1 frontier)
Top models
1
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is Phybench?
- PHYBench eval environment
- What is the current top score on Phybench?
- The top reported score is 30.0% by GPT-4.1 Mini, across 1 model reporting (1 from frontier labs).
- How can a model improve its Phybench score?
- Tools linked to Phybench on Sophon include Phybench RL Env (Prime Intellect) - RL environments, datasets, and scaffolds that target this eval.
- What license is Phybench under?
- Phybench is available under unknown.