AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
OpenCompass Hub
ZeroBench
ZeroBench is a challenging visual reasoning benchmark for LMMs. It consists of a main set of 100 high-quality, manually curated questions covering numerous domains, reasoning types and image type. Questions have been designed and calibrated to be beyond the capabilities of current frontier models.
Evaluation context and evidence
- Source
- OpenCompass Hub
- Publisher
- University of Cambridge, University of Alberta,etc.
- Benchmark release
- 2025-02-13
- Collected
- 2026-08-18
- Modality
- multimodal
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 0
This source record has no numeric results in our collection. That does not mean a score of zero or that no public results exist. Consult the original resources above.