Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Opus 4.8

anthropic.com

Post reply on HN
Print view View original

931–940 of 1001 posts

Re: Claude Opus 4.8

#931

Post by ulbu »

vlovich123 wrote: post #257
onlyrealcuzzo wrote: post #136

Earlier quoted context omitted.

I won't be surprised if the next gen frontier models are the last. There's orders of magnitude of low hanging juice to squeeze out of smaller models. It is almost guaranteed that a 60-90B model can outperform current SOTA in coding tasks within 2-3 years (design not certain, probably unlikely). It is far less clear that a 1.2T model will be meaningfully better enough to justify training it. As far as reasoning is con…

Took me a while to find what you were referring to by gram. Arxiv paper from 9 days ago that's not properly indexed by search engines. (G)enerative (R)ecursive re(A)soning (M)odels. They really wanted the acronym. https://arxiv.org/html/2605.19376v1

I propose GRIM: Generative Recursive Indeterministic Impression Machine.

Re: Claude Opus 4.8

#933

Post by artur_makly »

senko wrote: post #574
elAhmo wrote: post #485

Earlier quoted context omitted.

What is ultracode mode?

It's a combination of reasoning effort (max) + enabling workflow that orchestrates multiple sub-agents. After some interrogation, here's how it organized the work: 1. Design workflow (rts-game-design, 11 agents, ~13 min) ran first, produced SPEC.md + DESIGN.md: 1.1. Proposals (3 parallel agents): each designed a complete RTS from a different philosophy 1.2 Judge (1 agent): evaluated all three and synthesized one unif…

Just to confirm - you did not generate this plan/orchestration/harness - it did all that on its own?

Re: Claude Opus 4.8

#935

Post by blueblazin »

zuzululu wrote: post #377
thefounder wrote: post #373

>> As part of Project Glasswing, a small number of organizations are currently using Claude Mythos Preview Just f** off! I can’t wait for the Chinese models to catch up and bring these entitled as** holes down.

you mean after they scrape American LLMs ?

I think they poison outputs now if they detect distillation attempts. So a model trained on distilled outputs will be stupider.

Re: Claude Opus 4.8

#936

Post by cmrdporcupine »

hereme888 wrote: post #544

Early ArtificialAnalysis.ai results show GPT 5.5 is still the better bang-for-your-buck. OpenAI solves tasks with about 50% less output tokens. https://artificialanalysis.ai/?intelligence=coding-index&int...

This is true but OpenAI has been slowly boiling the frog here, too. $20 and $100 on their plan doesn't get you nearly as far as it did two, three months ago.

I use their (newish) 5x $100 plan and I routinely run out of weekly limits about a two days before the end of the week.

This has also goaded me into upgrading to $200 once before... and then had them hand out limits resets to everyone. Argh.

Re: Claude Opus 4.8

#938

Post by imtringued »

pseudosavant wrote: post #863
onlyrealcuzzo wrote: post #136

Earlier quoted context omitted.

I won't be surprised if the next gen frontier models are the last. There's orders of magnitude of low hanging juice to squeeze out of smaller models. It is almost guaranteed that a 60-90B model can outperform current SOTA in coding tasks within 2-3 years (design not certain, probably unlikely). It is far less clear that a 1.2T model will be meaningfully better enough to justify training it. As far as reasoning is con…

It is fascinating to me to see a new product category that improves so vastly year-after-year, where people commonly state that this is now the peak already. I couldn’t even imagine having to go back to a model from 12 months ago, much less 24 months ago. GPT-5.5 is so much better than GPT-4o that it sure seems like they keep finding new juice to squeeze. This is like going from dialup internet to DSL and acting like…

Yeah sure but is it so much better than Codex-GPT-5.3? No, if anything it's probably a little bit worse.

Re: Claude Opus 4.8

#939

Post by dilap »

jclay wrote: post #424
senko wrote: post #385

My fav coding benchmark for frontier models is to build a simple RTS game in one file (js/html/css). Claude Code with Opus 4.8 in ultracode mode nailed it, the best result so far: https://bsky.app/profile/senko.net/post/3mmwnrkwboc2v The prompt was: Create a simple but functional real time strategy (RTS) game similar to old WarCraft, StarCraft or Command & Conquer games. The player should be able to build buildings,…

It almost appears as if the code was minified. The variable names are short and formatting looks like it's written to minimize whitespace. Did it write it in this compact format all on it's own?

Doesn't look minified, just very dense, almost like progcomp code. First time I've seen an LLM spit out that style of code, I'm impressed!

Re: Claude Opus 4.8

#940

Post by lysecret »

simonw wrote: post #73

I generated pelicans riding bicycles on both thinking level low and thinking level high: https://gist.github.com/simonw/68560eddb0b268a8417f80ceb7304... The high one is notably better - the bicycle frame is the correct shape, unlike thinking level low. For comparison, here's Opus 4.7: https://gist.github.com/simonw/afcb19addf3f38eb1996e1ebe749c...

Sadly I think the correlation between this benchmark and performance is starting to break down imo. Still a legendary idea will be remembered and ingrained in the models forever haha
Post reply on HN