LVBench

A long-video understanding benchmark for retrieving and reasoning over information distributed across extended video inputs.

ModelsModels with a result.7
Top resultBest result on this benchmark.—
Top-3 spreadPoints from first to third.—
YearYear of release.—

Results

7 results
#ModelResultThe score from the source.
—Qwen3.8 Max81.8%
—LFM3-VL-2B41.2%