Skip to content

Benchmarks

Benchmark methodologies and vendor-reported results.

Entities

Entity Main site (canonical)
AutomationBench chinaaihub.com
BrowseComp chinaaihub.com
CyberGym chinaaihub.com
DeepSWE chinaaihub.com
GPQA Diamond chinaaihub.com
HLE chinaaihub.com
MMMU-Pro chinaaihub.com
SWE-bench chinaaihub.com
Terminal-Bench chinaaihub.com
Video-MME chinaaihub.com

Data policy

All fields follow the main site source policy: FACT / VENDOR CLAIM / ANALYSIS classification, official sources first, unknown where not publicly disclosed.