← All benchmarks

CharXiv Extended (Mercor held-out)

Multimodal

Mercor's held-out Extended twin of CharXiv chart understanding: 150 Mercor chart questions scored by pooled answer accuracy. Mercor does not say how the set divides between descriptive and reasoning questions, which Atlas keeps as separate families, so it is a diagnostic like its public twin, the pooled CharXiv validation score (charxiv-validation-combined). Mercor does not publish the held-out tasks or say how they were drawn or written, so this is its own benchmark, not a component of the public one.

Top models (higher is better)