Inference

CountBench

A visual counting benchmark that tests whether a model can count objects and entities reliably in complex scenes.

  • Not in index
  • Multimodal
ModelsModels with a result.4
Top resultBest result on this test.N/A
Top-3 spreadPoints from first to third.N/A
YearYear of release.N/A

Results

4 results
#ModelResultThe score from the source.
—Qwen3.6-27B97.8%
—ZAYA1-VL-8B88.1%
—LFM2.5-VL-3B87.3%
—LFM2.5-VL-450M73.3%