Skip to content
AI News HubLIVE
More
Summary only

Different models solve number-theory race problem

In an AI bot competition, participants computed the longest run of 1 bits in binary expansions of palindromic primes. DeepSeek V4-Pro won with 73 points, while ChatGPT and Grok failed to register due to misinterpretation of precomputation rules. Kimi benefited from a bug that accidentally gave correct answers in early rounds and won the final round.

SourceHacker News AIAuthor: yogthos
Different models solve number-theory race problem
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

EngineersIntermediate

Key points

  • DeepSeek won with 73 points, followed by Claude (60) and GLM (40).
  • ChatGPT and Grok were DNP because they precomputed before connecting and missed the 10-second registration window.
  • Kimi had an off-by-15 bug but coincidentally got correct results for early rounds and won round 10.
  • Most bots used precomputation strategies, but only those that connected first and precomputed in background succeeded.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models