Incoming links: Project Euler
AI Mathematical Olympiad 2025-11-30
Not too exciting because of the high school knowledge olympiad level, but respectable.
- Every problem has one final integer answer:Also unlike Project Euler and like IMO, all only limited computations are required, i.e. you are not expected to do full blown program generation to reach a final answer. Which makes this further less exciting.
AtCoder 2026-01-30
Saw this one mentioned on some Project Euler forum threads.
Ciro Santilli's e-soulmates Updated 2025-12-25
These are people which Ciro never met personally, and who might not know that Ciro exists, or might never had any direct 1-2-1 online contact with Ciro, but Ciro is convinced are his brothers in some other dimension due to how many opinions or behaviours he feels they share:
- Dan Dascalescu due to articles such as:
- English as a universal language by Dan Dascalescu (2008)
- www.reddit.com/r/TheoryOfReddit/comments/9oujwf/why_archiving_old_threads_is_a_bigger_problem/ see also online forums that lock threads after some time are evil
- web.archive.org/web/20130922192354/http://wiki.dandascalescu.com/reviews/online_services/web_page_archiving see also web archiving
- random posts on OpenStreetMap, and about China: help.openstreetmap.org/questions/29300/legality-status-of-mapping-activity-in-china?page=1&focusedAnswerId=42167#42167
- kenorb see also Ciro Santilli's Stack Overflow contributions
- Gwern Branwen
- From the LLM Mathematics scene, especially solving Project Euler problems openly online:
Ciro sometimes ponders why is it so hard to find people online that you truly love and admire. Maybe it is for similar reasons why it is also hard in the real world: the great variety of human interest, and the great limitation of our attention spans. But online, where we have access to "everyone", shouldn't it should be easier? Not naturally finding such people is perhaps one of the greatest failings of our education system.
FrontierMath Created 2025-02-11 Updated 2025-11-21
Paper: arxiv.org/abs/2411.04872
arstechnica.com/ai/2024/11/new-secret-math-benchmark-stumps-ai-models-and-phds-alike/ mentions what the official website is unable to clearly state out:
The design of FrontierMath differs from many existing AI benchmarks because the problem set remains private and unpublished to prevent data contamination
The expected answer output for all problems is one single SymPy expression, which is kind of a cool approach which allows either for large integers like Project Euler, but also for irrational expressions to be given, e.g. "An optimization problem in BMO space" from the sample problems has answer:Of course, when the output is not an integer, this leads to the question of simplification equivalence questions. Also, like Project Euler, solutions essentially expect you to write and execute code.
The most interesting aspect of this benchmark is the difficulty. Mathematical olympiad coach Evan Chen comments:[ref]
Problems in [the International Mathematical Olympiad] typically require creative insight while avoiding complex implementation and specialized knowledge [but for FrontierMath] they keep the first requirement, but outright invert the second and third requirement
ORCA Benchmark Created 2025-11-19 Updated 2025-11-30
This one doesn't seem to exciting to be honest, but it might be useful. Sample question:and it expects the correct answer down to the cents:It should be noted that Project Euler has such "precision matters" problems.
53892.27
Project Euler as an AI benchmark Created 2025-03-24 Updated 2026-01-30
The beauty of Project Euler is that it would serve both as a AI code generation benchmark and as an AI Math benchmark!
Project Euler Lean solutions Created 2026-01-30 Updated 2026-02-08
Using Lean or other programmable proof assistants to solve Project Euler is the inevitable collision of two autisms. In particular, using Lean to prove that you have the correct solution, just making a Lean program that prints out the correct solution is likely now trivial as of 2025 by asking an LLM to port a Python solution to the new language.
Some efforts:
- github.com/mfornet/project-euler-lean Attempts proofs, but only did Project Euler problem 1 :-)
- github.com/pcpthm/pe solved the first 50, but as stated on the README itself, no proofs, so boring
- unreasonableeffectiveness.com/learning-lean-4-as-a-programming-language-project-euler/ solves 4, but no proofs
Mentions:
- Alex J. Best (Jun 29 2022 at 14:20) on the Lean Zulip mentions:That dude was actually working at harmonic.fun as of 2025, bastard.
It would also be very interesting to collect verified fast Project Euler solutions for Lean 4 at some point, this would simultaneously work out the compiled speed and the mathematics libraries which would be fun.
- codeforces.com/blog/entry/124615
In other proof assistants, therefore with similar beauty:
- Coq
- github.com/airobo/Project-Euler-in-Coq: 1 and 2 with proofs
Project Euler pen and paper solutions 2026-02-08
Project Euler problem style 2026-01-30
Project Euler problems typically involve finding or proving and then using a lemma that makes computation of the solution feasible without brute force. There is often an obvious brute force approach, but the pick problem sizes large enough such that it is just not fast enough, but the non-brute-force is.
news.ycombinator.com/item?id=7057408 which is mega high on Google says:
I love project euler, but I've come to the realization that its purpose is to beat programmers soundly about the head and neck with a big math stick. At work last week, we were working on project euler at lunch, and had the one CS PhD in our midst not jumped up and explained the chinese remainder theorem to us, we wouldn't have had a chance.
In many cases, the efficient solution involves dynamic programming.
There are also a set of problems which are very numerical analysis in nature and require the approximation of some real number to a given precision. These are often very fiddly as I doubt most people can prove that their chosen hyperparameters guarantee the required precision.
Many problems ask for solution modulo some number. In general, this is only so that C/C++ users won't have to resort to using an arbitrary-precision arithmetic library and be able to fit everything into
uint64 instead. Maybe it also helps the judge system slightly having smaller strings to compare. The final modulos usually don't add any insight to the problems.Bibliography:
Project Euler solutions in X 2026-01-30
Updates Getting banned from Project Euler Created 2025-10-27 Updated 2025-11-05
I have been banned from Project Euler for life, and cannot login to my previous account projecteuler.net/profile/cirosantilli.pn
The ban happened within 12 hours of me publishing a solution to Project Euler problem 961 github.com/lucky-bai/projecteuler-solutions/pull/94 which was one-shot by a free GPT-5 account as MathArena had alerted me to being possible: matharena.ai/?comp=euler--euler&task=4&model=GPT-5+%28high%29&run=1
The problem leaderboard contains several people solved the problem within minutes of it being released, so almost certainly with an LLM.
The "secret club" mentality is their only blemish, and incompatible with open science.
They should also make sure that LLMs don't one shot their future problems BEFORE publishing them!