Skip to content

Benchmark

OmniVideoBench · Token reduction

Scores are percentages. Higher is better.

Models measured
1
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

48.8%
Mean
51.8%
Median
3.1%
Std dev
45.7%
10th percentile
51.8%
90th percentile
45.7%
Min
45.7%51.8%
The band covers the middle 80% of scores. The line marks the median.

Results

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

ModelScore
Qwen3.8 Omni FlashStatic → Agent mode · Qwen Code · official X post · 2026-09-18· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.8%
Qwen3.8 Omni FlashStatic → Agent mode · Qwen Code · release article · 2026-09-18· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.7%

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.