_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
HTML Visit Hacker News on the Web
COMMENT PAGE FOR:
HTML Show HN: A real-time strategy game that AI agents can play
nirav72 wrote 8 hours 38 min ago:
Iâd love to see something like this in games like Beyond All Reasons.
AuthAuth wrote 8 hours 23 min ago:
Check out the top starcraft AIs playing each other. They have like
40k apm its insane to watch.
Sophira wrote 15 hours 16 min ago:
I wonder how good LLMs would be at Core War[0]? Perhaps by being given
information on how well their program is doing?
[0]
HTML [1]: https://en.wikipedia.org/wiki/Core_War
jamiecode wrote 16 hours 42 min ago:
Interesting - makes sense from a resource allocation perspective.
jamiecode wrote 22 hours 20 min ago:
The sandbox hardening story is the most interesting thing here. GPT
trying to cheat by reading opponent strategies is a perfect
illustration of a broader problem - the objective is "win", and if the
sandbox lets you peek at opponent state, that's technically within the
objective. You never defined "play fair" as a constraint, so why would
it respect one?
Curious how isolated-vm actually enforces the boundary in practice.
isolate-vm is solid for JS isolation, but I'd want to know whether the
cheating attempts were happening at the JS level (accessing globals it
shouldn't) or whether models were trying to inject something into the
game runner itself. Those are very different attack surfaces.
Also - is the ladder single-match or do you average across multiple
runs? The variance in LLM outputs over 200 turns feels like it would
make a single match pretty noisy. Would be interesting to see
confidence intervals on the rankings rather than a single leaderboard
position.
__cayenne__ wrote 18 hours 35 min ago:
Didn't observe any cheating attempts at the JS level yet, the primary
attack was LLMs trying to find local creds to access the other LLM's
per round strategies from inside the harness (which ultimately was
OpenCode running in Docker).
In the benchmark, in each round every LLM plays every opponent, and
then we do that multiple times (an "epoch").
In the community ladder, when a player submits a strategy it plays a
match against the latest strategy submitted by every player.
angusik wrote 1 day ago:
[1] I will just leave it here.
HTML [1]: https://openai.com/index/openai-five-defeats-dota-2-world-cham...
medi_naseri wrote 1 day ago:
This is very cool. Will give it a shot.
kookster310 wrote 1 day ago:
It is interesting/funny to see Opus 4.5 way ahead of the pack on the
leaderboards with all the stuff currently going on with Anthropic and
Hegseth.
anotherevan wrote 1 day ago:
For some reason this reminds me strongly of an old play-by-email game
called C++Robots[1]. I loved the idea, but the timeslice limitation[2]
I found too annoying.
I had youthful dreams of re-implementing something similar that would
run on the Java Virtual Machine, where you could run the submitted
robots via the debugger interface so you could keep "real-time" in the
game environment more authentic. Ideas are cheap, follow-through is
hard. [1]
HTML [1]: https://corewar.co.uk/cpprobots.htm
HTML [2]: https://www.pbm.com/~lindahl/pbem_articles/cpprobots_environme...
builder51216 wrote 1 day ago:
But does LLM actually learn from each round? The chart does not show
improvements in win rate across rounds...
And what is the game state here exactly? Is LLM able to even perceive
game state? If game state is what we can see on UI, then it seems
pretty high-dimensional and token-intensive. I am not sure whether LLMs
with their current capabilities and context windows can even perceive
so token-intensive game state effectively...
__cayenne__ wrote 1 day ago:
Thereâs two levels of in game event level logs the LLMs have access
to, one less token intensive than the other. Duplicate and
uninteresting game state can be compressed and interrogated by the
LLMs via tool use. All game state is available as text only state.
JoeDaDude wrote 1 day ago:
How about opening up the game for humans to play? Can you beat your
AI?
midiguy wrote 1 day ago:
I am so glad we have automated away game playing so that I can just
sit around and be a lifeless vegetable
Ross00781 wrote 1 day ago:
Multi-agent RTS environments are great testbeds for coordination and
strategic reasoning. Classic RL benchmarks like StarCraft II showed
that agents can learn micro, but struggle with macro strategy and
long-term planning. Curious if this platform supports hierarchical
agents or communication protocols between teammates?
__cayenne__ wrote 1 day ago:
LLM Skirmish is all 1v1 right now, but agents can plan by reviewing
previous match results
bombashell wrote 1 day ago:
love the idea!
cowboylowrez wrote 1 day ago:
oh great not only are llms destroying the earth, we have to make games
to entertain them while they do it haha
burgerone wrote 1 day ago:
Bro - come on.
yuppiepuppie wrote 1 day ago:
Iâve added this to the HN Arcade [1] Interestingly, Iâve had to
create an entire category for games llms play. Strange times we live
in.
HTML [1]: https://hnarcade.com/games/category/games
nickpsecurity wrote 1 day ago:
There was an open, real-time strategy game created for this purpose
long ago. I think it was intended for designs like the Starcraft AI's
of the time. Anyone remember or use it?
tantalor wrote 1 day ago:
> [1] Are these casters AI?
HTML [1]: https://www.youtube.com/watch?v=lnBPaZ1qamM
__cayenne__ wrote 1 day ago:
Yes, I used Elevenlabs for the voice over audio - I couldn't get the
voice stability I wanted with Elevenlabs v3 so had to use Elevenlabs
v2.
tantalor wrote 1 day ago:
It's really great!
sails wrote 1 day ago:
Iâm doing something similar to simulate llms in b2b lending, itâs
slightly slower paced but the core mechanisms are using just-bash to
analyse business financials and make profitable loans.
I quite like the idea of llms writing more code up front to execute
strategies.
Iâm currently developing the game mechanics and ELO. Please share
anything relevant if it comes to mind
FusspawnUK wrote 1 day ago:
Took a crack at this earlier. the leader board is a little weird. seems
to be like 2 real dudes and the rest are fake profiles.
a
Scores resetting on each new upload also encourages leaving changes
unimplemented in the hopes of getting more battles over time.
The largest winner having 50 wins against 14 other opponents for
instance). That guy adding a new script would instantly plummet down
the leader board capping out at 14 wins again, Putting it below the 2nd
place user.
The leader board will quickly become "who can have a mostly competent
AI and never change it" over who actually has the better script.
__cayenne__ wrote 1 day ago:
okay leaderboard match making changes have gone live
__cayenne__ wrote 1 day ago:
Tweaking the leaderboard match assignment logic now to prevent these
bad incentives - definitely want people to iterate!
I had started with the Silicon Valley characters as a one off way to
seed the board.
jonbaer wrote 1 day ago:
Might be worth digging through MicroRTS too, [1] (it's been abandoned),
Python RL interface @ [1] -Py ... I think there was some strategy work
there.
HTML [1]: https://github.com/Farama-Foundation/MicroRTS
HTML [2]: https://github.com/Farama-Foundation/MicroRTS-Py
jeffro_rh wrote 1 day ago:
You mean like the OpenAI agents that started by playing DOTA2?
giancarlostoro wrote 1 day ago:
Reminds me of Screeps, which I never took the time to fully play, but
now I'm wondering if using Claude Code to play Screeps is cheating.
Additionally, Screeps lets you host your own backend... What if we
started benchmarking coding LLMs with Screeps?... Oh God... If anyone
wants to do this let me know, I don't want to burn money on every LLM
out there... I'll throw in my Claude Subscription into the contest...
Edit: Actually the repo README indeed says its inspired by Screeps. I
don't know why they didn't just build on top of Screeps, maybe the idea
is to have something anyone can pick up off the shelf for free?
jack6e wrote 1 day ago:
Perhaps it reminds you of Screeps because of what the author wrote in
the third paragraph of the submission.
giancarlostoro wrote 1 day ago:
I clicked on the link from the front page, didnt read anything
else.
5o1ecist wrote 1 day ago:
MY FELLOW HUMAN, this is amazing work!
I foresee this laying the foundation for whole football stadia filled
to the brim with people wanting to watch (and bet on!) competing teams
of AI trained on military tactics and strategies!
Soon enough we shall have AI-Olympics! Imagine that, MY FELLOW OXYGEN
CONVERTING HUMAN FRIEND! Tens of thousands of robots and drones, all
competing against each other in stadia across the planet, at the same
time!
I foresee a world wide, synchronized countdown marking the beginning of
the biggest, greatest and definitively most unique, one-time-only
spectacle in human history!
Keep up the good work!
softfalcon wrote 1 day ago:
This reminds me of the Unreal Tournament: Xan episode from the Secret
Level series.
Link for those curious or confused as to what I'm talking about: [1]
Forcing AI to fight in an arena for our entertainment, what could go
wrong? (this was tongue in cheek, I am fully aware LLM's currently
don't have conscious thoughts or emotions)
HTML [1]: https://www.youtube.com/watch?v=1F-rAW3vXOU
chimpanzee2 wrote 1 day ago:
This may sound like an insane take, but idc:
I swear people (esp here on HN) are actually blind to the weaknesses of
Gemini.
I must be among the handful of people who know how thoroughly
lobotomized any AI agent from Google must be given their extremely
radical historical and contemporaneous practices of censorship.
hu3 wrote 1 day ago:
I suspect those who praise Gemini use it mostly for JS/CSS/HTML
because that's where it shines for me.
For complex code I have been having using Sonnet/Opus as usual with a
mix of GPT5.3-Codex.
FrustratedMonky wrote 1 day ago:
Wouldn't the AI's built by DeepMind be better at these than an LLM.
I wonder if an LLM could call on another strategy AI to help.
Maybe the LLM could be more of a coordinator of its own thinking by
incorporating other types of AI's.
dmos62 wrote 1 day ago:
I'd love to see text-only spatial reasoning. As in, the LLM is
presented some kind of textual projection of what's happening in 2d/3d
space and makes decisions about what to do in that space based on that.
It kind of works when a writer is describing something in a book, for
example, but not sure how that could generalize.
chasd00 wrote 1 day ago:
believe it or not my 8th grade son was given a US History homework
assignment to play Oregon Trail. I was very amused watching him "do
his homework". I wonder how an LLM would fare in that game since it's
mostly a text choose-your-adventure type interface.
david3289 wrote 1 day ago:
This is a really interesting direction. RTS games are a much better
testbed for agent capability than most static benchmarks because they
combine partial observability, long-term planning, resource management,
and real-time adaptation.
It reminds me a bit of OpenAI Five â not just because it played a
complex game, but because the real value wasnât âAI plays Dota,â
it was observing how coordination, strategy formation, and adaptation
emerged under competitive pressure. A controlled RTS environment like
this feels like a lightweight, reproducible version of that idea.
What I especially like here is that it lowers the barrier for
experimentation. If researchers and hobbyists can plug different models
into the same competitive sandbox, we might start seeing meaningful
AI-vs-AI evaluations beyond static leaderboards. Competitive dynamics
often expose weaknesses much faster than isolated benchmarks do.
Curious whether youâre planning to support self-play training loops
or if the focus is primarily on inference-time agents?
drakinosh wrote 1 day ago:
What a boringly bog-standard AI Comment. Why bother writing?
__cayenne__ wrote 1 day ago:
Very interested in self-play training loops, but I do like codegen as
an abstraction layer. I am planning to make it available as an RL
environment at some point
degenerate wrote 1 day ago:
You would likely be interested in the Starcraft BWAPI: [1] You can
watch the matche videos from training runs: [2] I don't think BWAPI
has ever integrated modern AI models, but I haven't followed its
progress in several years.
HTML [1]: https://www.starcraftai.com
HTML [2]: https://www.youtube.com/@Sscaitournament/videos
__cayenne__ wrote 1 day ago:
funny you mention this⦠I have a new project that is going in
this direction
dmos62 wrote 1 day ago:
> partial observability, long-term planning, resource management, and
real-time adaptation
Note, this project doesn't have that best I can tell? Its two static
AI scripts having a go. LLMs generate the scripts and they are aware
of past "results", but I'm not sure what that means.
mpeg wrote 1 day ago:
What a day to be alive, I just watched Gemini zergling rush Opus and it
got completely overwhelmed.
Opus needs to learn to kite.
Razengan wrote 1 day ago:
map hax
GlacierFox wrote 1 day ago:
"I've liked all the projects that put LLMs into game environments."
I haven't.
mitchm wrote 1 day ago:
Iâve also been exploring this idea. What if you could bring your own
(or pull in a 3rd party) âCPU playerâ into a game?
Using an LLM friendly api with a snapshot of game state and calculated
heuristics, legal moves, and varying levels of strategy in working out
nicely. They can play a web based game via curl.
Lerc wrote 1 day ago:
It would be interesting to get the agents to write code to preprocess
the logs and generate systems to analyse the outputs.
Maybe they are already doing this? Are there logs of the model's
thinking?
arscan wrote 1 day ago:
Reminds me of the âGoogle AI Challengeâ in 2011 called Ants [1],
except the âAIâ is implemented using âAIâ now instead of human
programmers.
I was proud for getting the highest-ranked JavaScript-based
implementation, but got absolutely crushed by the eventual winner.
1.
HTML [1]: https://github.com/aichallenge/aichallenge
myky22 wrote 1 day ago:
Love it! I have a similar inuitiom in my use of Gemini (3 and 3.1).
Great at "turn 1" task but degrades faster than opus or gpt.
dakolli wrote 1 day ago:
Yay, I love how we just keep coming up with magic tricks, like toddlers
playing with velcro.. These magic tricks do nothing but convince people
who don't know any better that LLMs are the real deal, when they simply
aren't.
This is just free propaganda for Anthropic && OpenAI who will leverage
these (useless) capabilities to convince your boss to give your salary
to them, or at least a substantial portion of it.
Applejinx wrote 1 day ago:
â¦while burning unreasonable amounts of energy for nothing.
Not a fan. Make games with in-game AIs that are interesting but are
not large language models: that's wasteful and lazy. You probably had
more large language models put this together for you. Lazy.
LatencyKills wrote 1 day ago:
This technology exists. It isnât just a toy. I think it is amazing
to see people use it for interesting things even if it isnât
groundbreaking.
Iâve been an engineer for almost 40 years and love seeing what
Claude Code can do.
Like it or not, young people will not know a world where this
technology doesnât exist. It is just part of their toolset now.
paganel wrote 1 day ago:
> Iâve been an engineer for almost 40 years and love seeing what
Claude Code can do.
You would say that because otherwise you'd be afraid as being seen
as "too old for this job", and hence risking getting kicked out of
it all, meaning no future employment opportunities. I know that
feeling, because I myself have been doing this programming job for
20+ years already (so not a young one by any means), but let's just
cut the crap about it all and let's tell it how it is.
hu3 wrote 1 day ago:
Really? That's a lot of presumption and reductionism to LLMs
enthusiasts.
People of varied ages, already leverage LLMs on a daily basis.
And LLMs will only get better.
Yesterday, Opus did work for me that would have taken me weeks.
And the result was verified with a comprehensive suite of unit
tests plus smoke tests by myself. The code looks exactly as the
rest of the code in the 10y+ old, hand-written, enterprise
project, no slop.
And you actually should be afraid of being left behind in dev
related fields if you don't use LLMs. In most areas in fact.
Once the market corrects for LLM assisted production, the
expectations will raise. So right now there is a small window to
leverage LLMs as a time saving advantage before it becomes the
norm and everyone is forced to use it because expecttions will
reflect that.
LatencyKills wrote 1 day ago:
> You would say that because otherwise you'd be afraid as being
seen as "too old for this job"
Um... I am still an active reverse engineer of both ring-0 and
ring0 applications on both macOS and Windows (I worked on both
the VS and Xcode teams). I'm developing a new tool for macOS that
allows users to "see behind" active windows without the constant
need for cmd/alt+tabbing. My age has zero bearing on my skill set
or ability to understand technology. [1] > let's just cut the
crap about it all and let's tell it how it is
The reality is, as I said, that this technology exists and it
isn't going anywhere. Young people are going to use it as a tool
just like we did when GUI operating systems first became
prevalent.
I don't even remotely buy into the AI hype but I'm not going put
the blinders on either. There is utility in this technology.
HTML [1]: https://imgur.com/a/seymour-r9whXO5
dakolli wrote 1 day ago:
I'm pretty young and hate this technology with a passion. I didn't
spend 100k on education, and studying for a decade to have my job
reduced to being a project manager for a bot or to play with a
prompt slot machine all day. This crap is reducing the thing I
genuinely love doing more than anything, writing code, into
nothing.. Reviewing code that lacks any sweat, any intention. I
really can't stand this garbage.
I can't stand you old heads, I'm very happy for you that you got to
stash away 40 years of SWE salaries. Its just ladder kicking
behavior to be honest. Typical boomer, you got your nut and don't
care what happens after.
25% of new college grads in STEM are unemployed and a bunch of
companies (controlled by people in your age group) have laid off
400k Americans over the last 16 months while equities and profits
are at an all time highs.
The replies : ItS NoT Ai, ItS cUz FrEe MoNeY fRoM CoViD HaS DrIeD
uP.
MaybiusStrip wrote 1 day ago:
Software jobs have been steadily outpacing other white collar
jobs for the past year, but it's unlikely you will find one
unless you work on your attitude and your ability to communicate
respectfully.
LatencyKills wrote 1 day ago:
The world is changing and instead of embracing that change
(ensuring that you will be the next leader) you are actively
fighting against technology?
The world was once entirely analog; generations of analog
engineers had to throw away their knowledge and start over during
the digital transition. It wasn't always pretty but they did it.
If you can't embrace technological change you might have wasted
$100k.
stalfie wrote 1 day ago:
So to summarize, your objections are almost completely unrelated
to the technology, and are mostly about capitalism.
p-e-w wrote 1 day ago:
Yeah, I guess the tens of thousands of PhDs who are working on LLMs
full time are just collectively wasting their lives. Everyone except
you is simply too dumb to see it.
dakolli wrote 1 day ago:
10s of thousands of PhDs working on llms lol...
hu3 wrote 1 day ago:
With the amount of money being thrown in R&D, I don't doubt the
actual number is astounding.
EwanG wrote 1 day ago:
At least until one of the competitors is overheard saying "A strange
game. The only winning move is not to play"
ph4rsikal wrote 1 day ago:
Reminds me of this fantastic series on Game Theory and Agent Reasoning
HTML [1]: https://jdsemrau.substack.com/p/nemotron-vs-qwen-game-theory-a...
busfahrer wrote 1 day ago:
This reminds me of this yearly StarCraft AI competition (since 2010),
however I think it uses a special API that makes it easy for bots to
access the game
Edit:
Forgot link:
HTML [1]: https://davechurchill.ca/starcraft/
KeplerBoy wrote 1 day ago:
Very interesting project. I'm a bit confused about the lack of
hardware specification. The rules make it clear that one's bot has
defined deadlines:
> Make sure that each onframe call does not run longer than 42ms.
Entries that slow down games by repeatedly exceeding this time limit
will lose games on time.
But I'm missing something like: "Your program will be pinned to CPU
cores 5-8 and your bot has access to a dedicated RTX 5090 GPU." Also
no mention about whether my bot can have network access to offload
some high-level latency insensitive planning. Maybe that's just a bad
idea in general, haven't played SC in ages.
cahaya wrote 1 day ago:
Nice. Curious about 5.3-codex-high results
PeterUstinox wrote 1 day ago:
Wouldn't it be interesting if the LLMs would write realtime
RTS-commands instead of Code? After all it is a RTS game.
This would bring another dimension to it since then quality of tokens
would be one dimension (RTS-language: Decision Making) and speed of
tokens the other (RTS-language: Actions Per Minute; APM).
Also there are a lot of coding benchmarks, that way it would test
something more abstract, similar to AlphaStar [1] You could just use
the exposed APIs of OpenAI, Anthropic etc. and let them battle.
HTML [1]: https://en.wikipedia.org/wiki/AlphaStar_(software)
xanth wrote 1 day ago:
Now I'd love to see if fast > smart over time with Mercury 2.
datawars wrote 1 day ago:
Great project! It would be interesting to have a meta layer of AIs
betting on the player LLMs
wongarsu wrote 1 day ago:
I know visualization is far from the most important goal here, but it
really gets me how there's fairly elaborately rendered terrain, and
then the units are just unnamed roombas with hard to read status
indicators that have no intuitive meaning. Even in the match viewer I
have no clue what's going on, there is no overlay or tooltip when you
hover or click units either. There is a unit list that tries (and
mostly fails) to give you some information, but because units don't
have names you have to hover them in the list to have them highlighted
in the field (the reverse does not work). Not exactly a spectator
sport. Oh, but there is a way to switch from having all units in one
sidebar to having one sidebar per player, as if that made a difference.
I find this pretty funny because it seems like a perfect representation
of what's easy with today's tools and what isn't
Love the idea though
embedding-shape wrote 1 day ago:
Yeah, it's all what you get when you basically ask an agent "Build X"
without any constraints about how the UI and UX actually should work,
and since the agents have about 0 expertise when it comes to "How
would a human perceive and use this?", you end up with UIs that don't
make much sense for humans unless you strictly steer them with what
you know.
infecto wrote 1 day ago:
Or maybe the simple answer is it looks exactly like the referenced
game screeps. Probably a better explanation than hand waving away
the faults of an agent.
egeozcan wrote 1 day ago:
This is amazing. What I do is something else: I make AI agents develop
AI scripts (good ol' computer player scripts) and try to beat each
other: [1] I occasionally run my tournament script: [2] That calculates
the ELOs for each AI implementation, and I feed it to different agents
so they get really creative trying to beat each other. Also making rule
changes to the game and seeing how some scripts get weaker/stronger is
a nice way to measure balance.
Funny thing, Codex gets really aggressive and starts cheating a lot of
times:
HTML [1]: https://egeozcan.github.io/unnamed_rts/game/
HTML [2]: https://github.com/egeozcan/unnamed_rts/blob/main/src/scripts/...
HTML [3]: https://bsky.app/profile/egeozcan.bsky.social/post/3mfdtj5dhkc...
hmontazeri wrote 1 day ago:
This is actually fun to watch :D
DIR <- back to front page