Atlas

Benchmarks

← All benchmarks

ZeroBench Main (dots.vlm1 report)

Multimodal

ZeroBench main-question score reported as 4 in the original dots.vlm1 instruction-model card. The card does not state the dataset revision, sample count, pass@k definition, decoding settings or tool condition. Preserve this own-model reported value as a diagnostic rather than attaching an inferred pass@1 or pass@5 protocol.

Top models (higher is better)

ModelScore
dots.vlm1 Instruct4.0
Loading Atlas data…