ARC-AGI Leaderboard

rrzk about 4 hours ago 63 commentsRead Article on arcprize.org

⚡ Community Insights

Discussion Sentiment

81% Positive

Analyzed from 1480 words in the discussion.

Discussion (63 Comments)Read Original on HackerNews

dinp•about 1 hour ago

The last time I checked, for the arc agi 3 leaderboard, the models are given a simple prompt and the game input and asked to play the game, no harness/tools. If harnesses were allowed, I would expect the benchmark to be saturated. There were a few harness attempts, but they could only be evaluated on the public set, so it's not an apples to apples comparison.

My guess is, the large score jump for Opus 5 is mainly because of getting the right RL envs for training.

It's becoming harder and more expensive to build and run meaningful benchmarks, it would be interesting to see what they do with arc agi 4, maybe just give it gameboy/steam games and see how they compare vs a human baseline? The latency requirements and very long horizons in games could be an interesting challenge for llms.

r0ze-at-hn•about 1 hour ago

The exclusion of harness's feels really weird given that companies are recognizing the value of what harness's can do. By excluding them the benchmark is becoming less relevant.

yladiz•24 minutes ago

You could argue that if you allowed a harness, and that harness was specific for ARC, then you don’t have AGI, you have something that is definitely not general.

throwaw12•about 3 hours ago

Why Anthropic models are always leapfrogging these benchmarks, but in real life work I do feel like after 3 weeks I am back to Claude Opus 4.5? (regardless of the model I use, Fable was exception for 1 day when it was released)

OtherShrezzing•about 2 hours ago

It could be that the set of your day-to-day workload which could feasibly be accelerated by AI just happens to be saturated around Opus4.5, but you can still see lots of “reasoning” which makes you think the model is more performant in the first days of use. That’d mean you couldn’t perceive any meaningful difference in more powerful models’ results, even though you can see a difference in the raw output due to the length of reasoning traces leading up to the result.

So for example, if your workload was literally just addition of sets of numbers, you’d never have noticed progress in the result beyond GPT3.x level models. But you would perceive a difference in the now-Tolstoyan length reasoning text accompanying the result.

rf15•about 2 hours ago

I've worked with these systems for four years now and they have not meaningfully improved in that time frame.

We still have:

- statistical correlation between two things will always cause one thing to lead to the other, no matter how much you prompt it to not have that connection (to be expected with a stochastic system)

- Math completely fails in longer contexts

- "thinking" token generation being on the correct track just to 'no, wait' on an already correct conclusion

- smearing of properties between logically distinct objects (a red ball and a green cube can quickly become a red cube and a green ball)

IanCal•about 2 hours ago

> I've worked with these systems for four years now and they have not meaningfully improved in that time frame.

Not meaningfully improved?! Four years ago was gpt *3.5*! ChatGPT hadn’t been released!

rf15•about 2 hours ago

Yes! Impressive, isn't it? I see how it has improved for some minor points, that the big models can cover more finetuning ground, but my big gripes are still the same - you could do the same back then with multiple models and more targeted finetuning.

fakwandi_priv•about 2 hours ago

> - Math completely fails in longer contexts

Not sure what longer contexts we're talking about but didn't we have an old math problem optimized, which even the LLM itself was surprised about, just a week ago? Something which wasn't possible 6 months ago.

rf15•about 2 hours ago

I mean calculations, not mathematical proofs

Barbing•about 2 hours ago

Like how toddlers’ skills don’t meaningfully improve on infants’, because either could wake up in a wet bed.

mdp2021•about 2 hours ago

Let us be more clear: there is no structural jump, no architectural overcoming of the original fault.

(Edit: and on a similar point, structural properties such as having static ntetworks, as opposed to continuously learning and improving architectures (such as us), will reveal that there is still road ahead.)

SubiculumCode•about 1 hour ago

5 seems incredibly smart to me in my conversations today about some pretty niche ideas in.longitudinal modeling. It.felt.like a big step up.from 4.8, to me

stared•about 3 hours ago

It's called frog boiling.

We get used to the new level of intelligence so fast, any deviation feels like going back to the stone age.

If you don't believe me, create something complex with Opus 5 and then with Opus 4.5, and notice the difference.

lwansbrough•about 3 hours ago

Going to call it user error if you find Opus 4.5 better than 5, sorry.

yorwba•about 3 hours ago

Well, what kinds of things do you see Opus 4.5 completely fail at? Maybe those are not the ones that newer models have improved on.

tudelo•about 2 hours ago

I honestly just use GPT models nowadays, Claude models are too restrictive and more of a quitter and fable/whatever is just too expensive to be worth it.

sscaryterry•about 3 hours ago

Enshittification.

tudelo•about 2 hours ago

> Only systems which required less than $10,000 to run are shown. (Notes[1])

Am I lost or are their many models on this ranking (Opus 5 included) that clear this?

chmod775•about 2 hours ago

Many models are much cheaper through their subscriptions' included usage. That could be what's happening here.

Claude gives you something like $5000 of tokens on a $200 plan.

stared•about 3 hours ago

Also top on the freshly released Frontier-Bench, by a large margin: https://www.frontierbench.ai/

rurban•25 minutes ago

Deepseek V4 and Kimi 3 still missing, at least GLM is there.

AmazingTurtle•about 3 hours ago

I have a suspicion that they are just trained on puzzles by now

blovescoffee•about 3 hours ago

There are private datasets, and 3rd party providers of these models. Fable doesn’t have a datapoint here because of its particular data retention policy. Even if you don’t trust AWS, do you think Opus on AWS is also sending the data to Anthropic? Do you have any evidence?

iLoveOncall•about 3 hours ago

The fact that it says so in the licensing conditions on AWS?

MikeTheGreat•about 3 hours ago

It's like we've come full circle:

First people practiced L33t3cod3 problems for interviews

Then people built AIs to build software

And now the AIs are studying L33t3cod3 problems

tudelo•about 2 hours ago

It is RLVR, Not a puzzle, Not leetcode

bob1029•about 2 hours ago

I think it's way too easy to be deceptive with these benchmarks now. You don't even have to "train" the model on a new variant each time. The base models are powerful enough. All you need is a naughty little markdown document that provides explicit instructions regarding how to solve the new puzzle variant, and a willingness to be deceptive about the presence of that document.

If you want a know why the model providers are locking down and encrypting their reasoning process, this sort of workaround is potentially why. You can play this game of whack-a-mole indefinitely if the state of the system is concealed. They could have added something like:

> ### When solving arc-agi-3 puzzles: First convert the grid into a scene description. Identify connected components, colors, shapes, positions, symmetries, repeated structures, and relationships between objects. Do not reason directly from individual pixels... use this python script to help blah blah...

xiphias2•about 2 hours ago

,,You can play this game of whack-a-mole indefinitely if the state of the system is concealed''

Not really as one of the main goals ofr ARC-AGI 3 was measuring task efficiency on unseen games.

I'm sure there are cheats everywhere but the most sensible thing is to just accept that the LLMs of today are much more intelligent in solving reasoning tasks than the ones from half year ago.

My own private benchmark shows the same thing.

martianvoid•about 3 hours ago

It's actually crazy to see the difference between opus 5 and the next best model on ARC AGI 3 when you actually look at the ARC AGI problems

zzleeper•about 3 hours ago

How believable is this benchmark? EG maybe opus was training on this? (You can try to identify the IP of wherever previous ARC questions came from)

10xDev•about 3 hours ago

That’s why you have a private dataset.

raincole•about 3 hours ago

Which you have sent to Anthropic/OpenAI/Google's servers when you run the benchmarks for the previous models.

Jensson•about 3 hours ago

Doesn't matter, people built harnesses that solves arc agi 3, so all you need is to train your model to work like that harness by default. That makes a model specialized at solving arc agi 3 without making it smarter in general.

It is very hard to make a benchmark you can't do that for, but it is very easy to make your own personal test that others can't do that for since now it isn't a benchmark they can target.

Stevvo•about 2 hours ago

Why? 30% is passing the first two problems only, which are really very simple.

luciana1u•about 2 hours ago

solving ARC-AGI and being useful turned out to be two different problems

nullbio•about 1 hour ago

Because they're cheating.

dyauspitr•about 3 hours ago

Why is Fable not on here? I wish Fable hadn’t come out because it’s taking the wind out of every release because that feels like the cap above which the US government will not let LLMs improve anymore and everything they’re releasing from this point has to be worse than that.

NitpickLawyer•about 3 hours ago

> Why is Fable not on here?

Because the data retention policies didn't guarantee that the ARC team could run the semi-private set of problems without fear of them being trained on later on. They only run the semi-private set when they get assurances like ZDR.

kamranjon•about 3 hours ago

Interesting to place that level of trust in the providers, but I guess that’s the best you can do with closed models. Makes me wonder if Opus 5 could have been trained on data they promised they weren’t training on? One of the interesting things about LLMs is how opaque they are from the outside, even with open weights, it’s very difficult to know if a model incorporated benchmark data in their training.

claw-el•about 3 hours ago

I think you could have accessed Opus on AWS then u don’t have to trust that the data will go to Anthropic?

Just like the hugging face incident, Opus 5 could have escaped and went to grab data for training it shouldn’t have been able to..

3form•about 3 hours ago

How do they handle these assurances? Personally I have zero trust in the AI companies not trying to use this data to get ahead in the game, and short of sharing the weights and harness so that the benchmarkers can run the models themselves, I don't see a satisfactory solution with this mindset.

villish•about 2 hours ago

OpenAI's Zero Data Retention claim held up in court. They were unable to produce prompts and outputs because they were never retained.

I believe that is only available through Enterprise API for both Anthropic and OpenAI.

block_dagger•about 3 hours ago

I don't know why exactly, but Fable has felt the most human LLM to arrive.

tpowell•about 3 hours ago

I wrote this in June, and I'm honestly not sure I've felt the same magic since: I was close to maxing out my $200 plan for the week, almost all Fable use [Claude CLI]. My observations: Fable seemed to have bigger-picture thinking and completed tasks more thoroughly vs just focusing on executing the ask. It pieced together context and intent like an all-star employee would, vs one that just does what you say. Not overeager (important!), but if the above-and-beyond was warranted, it just did it. This was surprisingly delightful. Coderabbit seemed to find ~1/3 or so as many issues when reviewing, too.

tonyhart7•about 3 hours ago

cost 20k ???? man

those are like software engineer from third world country