back
56 comments
The "Diamond Sponsors" of the event are Huawei and OpenAI. I found the welcome note from OpenAI [1] quite curious:

"[...] Eventually, AI will be able to solve even the hardest contest problems that we’ve seen yet. It will work alongside us and help drive the discovery of new knowledge. What you take from this week - the sense of being stuck, the thrill of progress, and the practice of building together - will remain critical as you shape your community and the future you build. [...]"

By Chief Scientist of OpenAI, Jakub Pachocki. Who happens to be an incredibly accomplished Competitive Programmer (2nd in ICPC World Finals, Winner of Code Jam, 2nd in Hacker Cup).

[1] https://icpc.global/community/history/brochures/world-finals...

That curious statement comes across as inappropriate. These talented students are among the best in the world. Yet, openai chose to make it about and praise themselves. That was not classy.
Surely it's meant to reassure the contestants that they're not wasting their time. They'll obviously know AI can do some of this stuff so it's a reminder that the true purpose of the competition isn't to be good at programming.

People still compete in playing musical instruments, riding horses, painting pictures, etc. All redundant because of technology but still they do it for other reasons, not the practical utility of the product of their work.

Peak HN here. Playing musical instruments is "redundant because of technology". And so is "riding horses"? And "painting pictures". Eh, what?
Yes? I do not play the guitar to get a better recreation of a song - if that’s all I wanted I’d play the original on high quality speakers.
For example, almost all music listened to today is recorded, not live. It was the opposite way around 100+ years ago. Almost all portraits are photographs, not paintings, etc. The common practical uses of those skills are gone and they're relegated to small niches, luxury services, and personal pleasure.
Lol and the Huawei note wrote otherwise. AI still hallucinates a lot so pursue mastery over algorithms and data structures to improve the technology
Chinese century incoming.
How many chief scientists (aka career climbers) does OpenAI have at this point?
meret is really good tho.
I was onsite today watching the contest live, and great atmosphere all around. One surprising outcome: the team in 17th place solved the same number of problems as the team that won gold in 4th place. Hopefully that isn't too demotivating to any team and we can see better separation in the future. After all, it can only mean that the problemsetters underestimated the contestants ;)

Congratulations to all the teams!

> the problemsetters underestimated the contestants

Except for problem C, which was only submitted by 4 teams, all unsuccessfully.

I don't blame them. That problem statement seems to be deliberately confusing.
Yes, I meant to imply that the problemsetters are to blame.
I doesn't seem that unclear to me? I absolutely don't have the skill to solve it, but it took <2 minutes to understand the problem and goal.
It doesn't seem that hard to solve to me either. It's solvable with basic linear programming.

    1. Add a variable for each node, and a variable for each output edge from stations.
    2. For each reservoir add equality constraints to the sum of incoming edges with the coefficients given in the problem.
    3. For each station add equality constraints between the weighted sum of its inputs (which is 1 for the root station) and its outputs (which are the variables we added).
    4. Add an out_edge >= 0 constraint for each output edge on stations to forbid illegal negative flows.
    5. Add a variable m which is constrained to be less than all the output station variables.
    6. Maximize m.
The headers A-L are links to the problem PDFs if you want to see.
The ICFP progamming contest for this years International Conference on Functional Programming started today if anyone's interested.

https://icfpcontest2025.github.io/index.html

In the TV show Silicon Valley, there’s a joke that Nelson “Big Head” Bighetti, the perennial underperformer, did his undergrad at ASU. But I guess that’s one thing the show got wrong, because Arizona State University finished among the top three American schools in the world finals.
Great memories of ICPC. Their problem statements are so creative it’s insane.

Nothing like CF garbage, solving imaginary patterns that no one cares about.

What do you mean by imaginary patterns from Codeforces?
The problems are really low quality.

Topcoder/ICPC/CodeJam have the best problem statements.

Check out this for example: https://codeforces.com/contest/2136/problem/C

Why would I care if some array can be obtained by combining some other patterns?

This is a pretty neat problem. Maybe this isn't the activity for you...
No, sorry. You have a bad taste.
Wow ASU, the 4th highest US School. Not bad when also being a top party school.
You can check all the teams and their members here: https://cphof.org/advanced/icpc/2025

It's rarely the case that looking at school names is useful (for many things in life) when there are more data points.

In this case, without any insider knowledge, just by looking at their profiles, the relevant name would appear to be Benjamin Jeter (https://codeforces.com/profile/BenjaminJ) rather than ASU. Currently 5th active American in the top competitive programming platform, top 200 worldwide (https://codeforces.com/ratings/country/United%20States). That's elite.

In teams of 3, even one "super player" can make a big difference. Almost certainly carrying that team.

Does anyone know how long it will be until the input/output data packet is made available? I'm interested in taking a crack at some of these but know I always miss an edge case.
It has been made available here: https://open.kattis.com/problem-sources/ICPC%20World%20Final...

It does seem to be missing 3 problems though.

You can try your solutions here: https://wf25open.kattis.com/contests/wf25open

I believe it runs with the same test data as in the actual contest.

> You need to join the contest in order to be allowed to submit.

Doesn't seem to be publicly accessible to non-participants (yet?)

Wow, China with a pretty dominant position there with 3 teams in the top five. Interesting seeing Karlsruhe there too, I'd been looking at that school.
They have won the championship multiple times actually. There are times where "Western teams" are just full of mainland Chinese communicating in Mandarin (e.g. a few MIT teams in the past including the 2021 champions...)
Wowser. Is it a similar story to the Chinese olympics successes - train kids from a (very) young age in specialised schools?
My alma-mater is only 61st. I remember the times we’re top 30. Great atmosphere though! I wonder if these competitions will become obsolete as LLMs advance.
I don’t see why it would. Chess is still bigger than ever.
I expected fast response from AI labs about what kind of medals top models could win.
What is Moscow's secret to training their programmers so well?
*St. Petersburg

I guess, like a lot of other sports at the college level, having a reputation that attracts the best competitive programmers (and a great coach to go along with it) doesn't hurt: https://en.wikipedia.org/wiki/Andrey_Stankevich

The "Russian hacker" stereotype is not at all fiction. If you've ever worked with developers from any of the former USSR, you'll know what I mean; they are not afraid to be creative and do whatever it takes to solve the problem.
Start training them from a very young age: competitive coders typical start in secondary/high school or even earlier.

Russia has specialized schools to support this kind of education, so are China and many other countries. Thus, they rank very high in IOI and ICPC contests

I expect OpenAI to have crushed the competition here.
what's the meaning of of the A,B,C... columns It's a ratio but of what over what ?
A, B, C... are the ID of the problems (not sorted by difficulty, teams have to figure out the order).

For each cell of (team, problem), the "ratio" x/y isn't actually a ratio. x means how many submissions were made. y is the minute when the accepted solution was submitted. For example, "3/298" means they made 3 submissions in total, and the accepted solution was submitted in the 298th minute (the contest lasted 300 minutes).

The penalty ("Time" column) is calculated as (sum of y) + 20 * (sum of (x-1)).

The columns are the questions. The score is “total att/solved”, (attempted/ solved?) though that doesn’t help much. I’m guessing you need to get the answer within the time limit.

https://worldfinals.icpc.global/scoreboard/2025/problems/A.p...

They're the problems with how many teams correctly solved it over the number of teams that attempted to solve it.