Skip to content

Benchmark

HLE-VL (w/ Tools)

Scores are percentages. Higher is better.

Models measured
1
Scores recorded
1
Sources
1
IQ weight
No weight published

Score distribution

52.2%
Mean
52.2%
Median
0%
Std dev
52.2%
10th percentile
52.2%
90th percentile
52.2%
Min

Leaderboard

#ModelScore
1Qwen3.8 MaxQwen3.8-Max release vision/video table · Code Interpreter + Search· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.52.2%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.