@cline: The International Math Olympiad is the hardest math competition in the world, where students compete on proof problems …
Summary
The International Math Olympiad is the world's hardest math competition, and DeepSeek V4 Flash, an AI model, won a gold medal for only 12 cents, showcasing impressive cost-effective performance.
View Cached Full Text
Cached at: 08/27/26, 03:28 AM
The International Math Olympiad is the hardest math competition in the world, where students compete on proof problems most math PhD’s even struggle with.
DeepSeek V4 Flash won a gold medal for only 12 cents. https://t.co/DAV3kGPIco
Similar Articles
@cline: DeepSeek V4 Flash won a gold medal on the IMO and almost 2x'd the human median score. Free in Cline: 1. npm i -g cline …
DeepSeek V4 Flash achieved a gold medal at the International Math Olympiad, almost doubling the human median score, and is available for free through Cline.
MIT scientists build the world’s largest collection of Olympiad-level math problems, and open it to everyone
MIT researchers, in collaboration with KAUST and HUMAIN, have released MathNet, the largest open-source dataset of Olympiad-level math problems, containing over 30,000 expert-authored problems from 47 countries.
@ChrisHayduk: https://x.com/ChrisHayduk/status/2076196217109746095
This article compares two AI approaches for mathematical problem solving: DeepMind's AlphaProof, which uses reinforcement learning in the Lean proof language, and OpenAI's raw LLM that achieved a gold medal at the 2025 International Math Olympiad without formal methods.
MIT & the IMO released MathNet, the world’s largest dataset of International Math Olympiad problems & solutions. MathNet is 5x larger than previous datasets & is sourced from over 40 countries across 4 decades
MIT and the IMO release MathNet, a massive dataset of International Math Olympiad problems and solutions spanning 40 years and 40+ countries, 5x larger than prior datasets.
Solving (some) formal math olympiad problems
OpenAI achieved a new state-of-the-art 41.2% on the miniF2F formal math olympiad benchmark using a technique called 'statement curriculum learning,' which iteratively trains a neural prover on proofs of increasing difficulty. The approach builds on iterative proof search and retraining over 8 iterations to significantly outperform the previous best of 29.3%.