MAGENTA reportedly solves all six IMO 2026 problems with K2-Horizon-7B
MAGENTA’s announcement describes a feedback loop between natural-language reasoning and formal verification. It claims 100% accuracy across 93 AIME and HMMT problems using K2-Horizon models of different sizes.
TLDR
According to MAGENTA’s announcement, the pipeline achieved 100% accuracy across 93 problems from AIME 2025, AIME 2026 and HMMT February 2026 using K2-Horizon models of different sizes. The announcement also says MAGENTA paired with the compact K2-Horizon-7B solved all six IMO 2026 problems. Those solutions, it says, were assessed using automated grading with reference answers and separately reviewed by IMO medalists.
Combined views
5
1 Source, first seen ago