Claude Breaks New Record on the Riemann Hypothesis Benchmark, Report Says Model Is Unannounced

A Claude model has set a new record on a benchmark built around the century-old Riemann hypothesis, pushing the lower bound of verified cases substantially higher, according to Chinese outlet QbitAI.

The report says the result appears to come from an unannounced new model, hinting that Anthropic may have a stronger reasoning model in internal testing.

The Riemann hypothesis is one of mathematics' most famous open problems, first posed more than a century ago, concerning the distribution of the nontrivial zeros of the Riemann zeta function. On this kind of benchmark, the goal is not a simple answer but pushing the interval over which the hypothesis can be computationally verified ever further outward.

QbitAI describes the new record as pushing the lower bound up by a large margin — a concrete mathematical milestone rather than a gain on a conversational or common-sense benchmark.

The signal matters because frontier models are increasingly involved in mathematical research; setting a record on a century-old problem suggests real progress in long-horizon reasoning and formal verification.

The open question is whether Anthropic will confirm the model, and whether the approach generalizes to other open problems in number theory.