NearAI Claims to Solve 672 Math Problems for $111

NearAI has claimed to solve a total of 672 Lean 4 problems from the mathematical formal proof benchmark PutnamBench for $111. This case could demonstrate the cost efficiency of AI agents. NearAI's Lean agent solved all questions in the PutnamBench, reportedly achieving a cost that is 250 times lower than competing systems. However, it is not clearly stated whether the conditions for comparison, such as model, execution budget, search method, and retry structure, are the same. The currently confirmed fact is that NearAI completed 672 problems and presented a total cost of $111. It is not yet reflected independently on the public leaderboard of PutnamBench. The official PutnamBench repository presents the number of problems based on Lean 4 as 672, and formal proofs are evaluated based on whether the AI provides verifiable proof procedures. NearAI's claim is significant within the NEAR ecosystem, which emphasizes 'verifiable AI.' Trusted execution environments are technologies that process data in isolated hardware environments to restrict external access. In February, NearAI highlighted the infrastructure for agents to compute and transact, and this claim adds to the case for cost efficiency. Competition in the field of formal proofs is ongoing, with MistralAI reporting that LeanStral 1.5 achieved 587 out of 672 on PutnamBench, presenting a cost of about $4 per problem. The PutnamBench leaderboard is discussing better ways to represent compute budgets by access type. The impact of NearAI's performance on NEAR prices or ecosystem demand remains uncertain.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.