Skip to content

Benchmark checker Maths

Is MathArena still meaningful?

Still useful Checked against the report, 5 November 2026

Yes. Fresh contests limit contamination, but small contests give wide error bars.

Why still useful: The chapter rates it Active: fresh contests keep the problems new, and the top score is 84.0%.^{[1][1] MathArena Primary}

What “Still useful” means

Known problems

Each contest is small, so error bars are ±2 points or more.^{[1][1] MathArena Primary}

What to look at instead

The report names no direct successor.

Where it appears in the report

Chapter 1, Benchmarks

Table 5. Reasoning and knowledge benchmarks: what they measure, who maintains them, the top score on 5 November 2026 and the known flaw in “Go deeper: Reasoning and knowledge benchmarks (17)”

Sources

Copied from the chapter source lists. Each number links back to the source in its chapter.

[1]
MathArena. It scores models on fresh maths competitions. Primary matharena.ai Chapter 1, source 33