NEAR AI Lean Agent Solves Putnam Bench Problems
Post claims low-cost completion of all benchmark tasks by NEAR system.
TLDR
Lukasz Kaiser retweeted a message from Alex Skidanov on social media. The message states that the NEAR AI Lean agent solved all problems in Putnam Bench. It achieved this at 1/250 the cost of the second cheapest submission. The total came to $111 for 67 submissions according to the post. Kaiser is identified as a senior research scientist at OpenAI and co-author of the Transformer paper. The post is shared in the context of AI research discussions. The information reflects what was stated in the retweeted content without further verification from additional sources in the available posts.
Combined views
558
2 Sources, first seen ago