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

771–780 of 1001 posts

Re: Claude Opus 4.8

#771

Post by rphv »

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?

"Readability by humans" may no longer be as important as it once was.

Re: Claude Opus 4.8

#772

Post by gertlabs »

We just finished our initial coding evals of Opus 4.8. Anthropic definitely heard the backlash from Opus 4.7 and they made up for it today.

Subjectively, it's also quite enjoyable to use (although it feels a bit slower on max reasoning), and it's the first Anthropic model that can implement a complex feature without Codex finding 100 bugs.

Data at https://gertlabs.com/rankings

Re: Claude Opus 4.8

#773

Post by silvertaza »

Frannky wrote: post #736

I use 4.6, because 4.7 is super lazy, deflects responsibility, and assumes it is good and I am bad, and avoids checking reality. It looks like it's trained on lazy humans instead of good engineers. Should I try 4.8? I am happy with 4.6. I am not happy with 4.7.

I have the exact same experience, word-for-word. I'm fascinated not everyone sees that.

Re: Claude Opus 4.8

#775

Post by ElkeQin »

NiloCK wrote: post #77

A rambling comment: I think this is the first time we've had a third minor version bump on a frontier Anthropic model. (I count the 0.5s as major here, because they've been issued non-sequentially and also corresponded to massive capability leaps, eg, Sonnet 3.5, Opus 4.5). So now the Opus 4.5 family has successors 4.6, 4.7, and 4.8, each posting fairly modest claimed gains. My own experience w/ 4.6 and 4.7 are that…

[flagged]

Re: Claude Opus 4.8

#776

Post by byzantinegene »

jimbokun wrote: post #447
NiloCK wrote: post #77

A rambling comment: I think this is the first time we've had a third minor version bump on a frontier Anthropic model. (I count the 0.5s as major here, because they've been issued non-sequentially and also corresponded to massive capability leaps, eg, Sonnet 3.5, Opus 4.5). So now the Opus 4.5 family has successors 4.6, 4.7, and 4.8, each posting fairly modest claimed gains. My own experience w/ 4.6 and 4.7 are that…

How long would it take to evaluate a new coworker to say “wow she’s really bright?” Relative to your other coworkers? A few days? A few weeks? Longer? However a company releases a new AI model and within hours users are confidently proclaiming how much smarter it is than previous versions.

alot of investor money is hinging on models performing better every release.

Re: Claude Opus 4.8

#777

Post by gertlabs »

swingboy wrote: post #718
gertlabs wrote: post #293

Earlier quoted context omitted.

4.5/4.6 were roughly the same in our testing. Opus 4.7 is smarter, but it's difficult to use as a product for various personality issues. So far, Opus 4.8 seems to be going down that path (unusably slow, but this could be a launch day rollout problem). Full Opus 4.8 tests are in progress now. Data at https://gertlabs.com/rankings

Looking forward to the results. Thanks for your work.

Appreciate that! Results are live: https://gertlabs.com/rankings

Opus 4.8 is the first tangible improvement since Opus 4.5. And it doesn't seem to have the personality problems of the last release -- I've been enjoying using it.

Re: Claude Opus 4.8

#778

Post by DoctorOetker »

Gomotono wrote: post #405
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…

I don't think this is true at all. It might feel like this because we are used to a very very fast release cycle but we are only in this topic for a few years. We have so many ways of optimizing: - continusly creating more and better training data - increasing parameters to 20/50/100TB - We still wait for Mythos access - We still wait for Mythos distilation (i haven't heard any rumors or so that there is a distilled…

> There was also a research paper were they showed that a LLM can compute things.

Can you be a little more specific than that or provide a reference?

I assume you're not indicating universality of neural networks?

Re: Claude Opus 4.8

#779

Post by Spikefu »

Spikefu wrote: post #752
Spikefu wrote: post #727

I was happily plodding away with it earlier when it threw this out in the middle of a response in Claude code: --- So — what did you actually see before you hit Ctrl-C? That's the信号 I'm most curious about, and it tells us what to --- That's the sort of behavior I'd expect from a one or two year old model quantized down to about 1 bit - right word, wrong language in a response. Google translate tells me that's Chinese…

Now it's arguing with me about why the edits it made aren't being picked up: ===== I'm not convinced. I didn't run it until you told me it was all fixed up. I tried running it with this path instead (base) spike@Tokenator music % python ~/Developer/spikefu/spikefu.py and it worked. It also now works with the old relative path approach. You're right, and I owe you an honest correction — I've now given you two confiden…

It did one more weird thing where it seemed to lose the ability to count in order: =====

You're right that hard per-action gating causes loops. So rather than block the model, make the harness own the critical structure deterministically, leave a git trail, and use soft guidance:

- (2) Commit breadcrumbs: /new commits the scaffold immediately (before discovery), then commits discovery progress. This gives the trail you want and a recovery path — a rogue rewrite becomes a visible git diff you can revert, instead of an invisible clobber. This is better than gating for your concern. - (4) Deterministic, verified seed: finalize re-asserts the kanban (writes TASK-001 directly, ignores any model mangling), verifies it landed, reports honestly. The harness guarantees structure regardless of what the model did. - (3) Soft-constrain discovery: prompt says "the .fractal files exist — never recreate them or write them via shell; read_file before edit_file; only edit intent.md/constraints.md." Guidance, not a hard gate. - (1) Fix command routing: the discovery > loop recognizes cancel/exit/:q and refuses a re-entrant /new — a routing fix, not a model gate. This alone would have prevented Run 2's disaster.

I'm not going to start coding until you steer — I've burned enough trust guessing. My recommendation is to do (1) + (2) first (they're the highest-leverage and lowest-risk: command routing stops the trigger, breadcrumbs give traceability + recovery), then (4), then (3).

=====

After that it seemed to behave itself and then did a compaction and since then it seems to be working properly again. Very odd. (and disconcerting)

Post reply on HN