Frontier AI models outperform junior accountants on benchmark tasks
Read original source →mercor.com
Frontier AI models now surpass junior accountants in speed and accuracy on medium-length, well-defined accounting tasks. Eighteen months ago, the best models scored below the average accountant's 37% benchmark performance.
2 sources
More stories today
Coding Agent Index adds safety refusal reporting
Artificial Analysis·4 hours ago
GLM 5.3 and GLM 5.3 Flash land in Cursor
Cursor·4 hours ago
Hugging Face Hub to let users send community policies to Microduck robots
Victor M·4 hours ago
MiniMax H3 video generation extended to 30 seconds
MachGen AI·4 hours ago
Researcher: LeWM's per-timestep SIGReg allows temporal collapse
Randall Balestriero·4 hours ago
Acemoglu questions whether AI boom can continue without worsening inequality
Daron Acemoglu·4 hours agoDeveloper builds first OpenAI ModRetro Chromatic game
Lizzie Siegle·4 hours ago
開発者がM5StackでDots向け携帯デバイスを自作
あいろぐ·4 hours ago