Clockbench
Frontier
ClockBench: multimodal clock reading and reasoning benchmark implemented for verifiers.
- Domain
- rl-env
- License
- unknown
- Published
- Mar 2026
Cite
Notes
Only stored in your browser.
Top score 40.0% by GPT-4.1 Mini - 5 models reporting (5 frontier)
Score history
5Top models
5
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is Clockbench?
- ClockBench: multimodal clock reading and reasoning benchmark implemented for verifiers.
- What is the current top score on Clockbench?
- The top reported score is 40.0% by GPT-4.1 Mini, across 5 models reporting (5 from frontier labs).
- How can a model improve its Clockbench score?
- Tools linked to Clockbench on Sophon include Clockbench RL Env (Prime Community) - RL environments, datasets, and scaffolds that target this eval.
- What license is Clockbench under?
- Clockbench is available under unknown.