HNGlance

Top Best New Show

I compared my daughter against SOTA models on math puzzles

https://blog.michalprzadka.com/posts/gmil-benchmark/

by przadka • 1 year ago

15 3 1 year ago

BL

blog.michalprzadka.com

Related Stories

Loading related stories...

Comments

3

przadka• 1 year ago

I tested o3, r1, 4o and other SOTA models against puzzles from an international math competition and compared their performance with my 11-year-old daughter's solutions. Full results include detailed conversations with each model and complete methodology.

michalwarda• 1 year ago

Very cool post! I wonder how much will it affect the psychology of next generations.

ukituki• 1 year ago

Interesting how the reasoning differs between models, e.g. DeepSeek trying the brute force tricks

Source preview

blog.michalprzadka.com

HNGlance

© 2024-2026 HN Glance. Story content belongs to its original sources.

Privacy
Terms