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

Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

tokens.billchambers.me

Post reply on HN
Print view View original

401–410 of 620 posts

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#401

Post by nothinkjustai »

tossandthrow wrote: post #298
nothinkjustai wrote: post #280

Earlier quoted context omitted.

Who cares? Why are people so invested in trying to “convert” others to see the light? Can’t you be satisfied with outcompeting “non believers”? What motivates you to argue on the internet about it? Deep down are you insecure about your reliance on these tools or something, and want everyone else to be as well?

Why do people invest themselves so hard in interjecting themselves into conversations about Ai telling people it doesn't work? It feels so off rebuilding serious SaaS apps in days for production, only to be told it is not possible?

Who here said ai “doesn’t work”?

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#402

Post by axpy906 »

rectang wrote: post #68

For now, I'm planning to stick with Opus 4.5 as a driver in VSCode Copilot. My workflow is to give the agent pretty fine-grained instructions, and I'm always fighting agents that insist on doing too much. Opus 4.5 is the best out of all agents I've tried at following the guidance to do only-what-is-needed-and-no-more. Opus 4.6 takes longer, overthinks things and changes too much; the high-powered GPTs are similarly f…

Why not just use Sonnet?

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#403

Post by braebo »

danny_codes wrote: post #251
Bridged7756 wrote: post #233

Earlier quoted context omitted.

Mirrors my sentiment. Those tools seem mostly useful for a Google alternative, scaffolding tedious things, code reviewing, and acting as a fancy search. It seems that they got a grip on the "coding LLM" market and now they're starting to seek actual profit. I predict we'll keep seeing 40%+ more expensive models for a marginal performance gain from now on.

I just don’t see how they’ll be able to make a profit. Open models have the same performance on coding tasks now. The incentives are all wrong. Why pay more for a model that’s no better and also isn’t open? It’s nonsense

Which open model has the same performance as Opus 4.7?

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#404

Post by QuantumGood »

andai wrote: post #239

For a fair comparison you need to look at the total cost, because 4.7 produces significantly fewer output tokens than 4.6, and seems to cost significantly less on the reasoning side as well. Here is a comparison for 4.5, 4.6 and 4.7 (Output Tokens section): https://artificialanalysis.ai/?models=claude-opus-4-7%2Cclau... 4.7 comes out slightly cheaper than 4.6. But 4.5 is about half the cost: https://artificialanalysi…

Some have defined "fair" as tests of the same model at different times, as the behavior and token usage of a model changes despite the version number remaining the same. So testing model numbers at different times matters, unfortunately, and that means recent tests might not be what you would want to compare to future tests.

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#406

Post by rectang »

axpy906 wrote: post #402
rectang wrote: post #68

For now, I'm planning to stick with Opus 4.5 as a driver in VSCode Copilot. My workflow is to give the agent pretty fine-grained instructions, and I'm always fighting agents that insist on doing too much. Opus 4.5 is the best out of all agents I've tried at following the guidance to do only-what-is-needed-and-no-more. Opus 4.6 takes longer, overthinks things and changes too much; the high-powered GPTs are similarly f…

Why not just use Sonnet?

I've used Sonnet a lot. It is not as good as Opus at understanding what I'm asking for. I have to coach Sonnet more closely, taking more care to be precise in my prompts, and often building up Plan steps when I could just YOLO an Agent instruction at Opus and it would get it right.

I find that Opus is really good at discerning what I mean, even when I don't state it very clearly. Sonnet often doesn't quite get where I'm going and it sometimes builds things that don't make sense. Sonnet also occasionally makes outright mistakes, like not catching every location that needs to be changed; Opus makes nearly every code change flawlessly, as if it's thinking through "what could go wrong" like a good engineer would.

Sonnet is still better than older and/or less-capable models like GPT 4.1, Raptor mini (Preview), or GPT-5 mini, which all fail in the same way as Sonnet but more dramatically... but Opus is much better than Sonnet.

Recent full-powered GPTs (including the Codex variants) are competitive with Opus 4.6, but Opus 4.5 in particular is best in class for my workflow. I speculate that Opus 4.5 dedicates the most cycles out of all models to checking its work and ensuring correctness — as opposed to reaching for the skies to chase ambitious, highly complex coding tasks.

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#407

Post by krackers »

_blk wrote: post #164
hgoel wrote: post #74

The bump from 4.6 to 4.7 is not very noticeable to me in improved capabilities so far, but the faster consumption of limits is very noticeable. I hit my 5 hour limit within 2 hours yesterday, initially I was trying the batched mode for a refactor but cancelled after seeing it take 30% of the limit within 5 minutes. Had to cancel and try a serial approach, consumed less (took ~50 minutes, xhigh effort, ~60% of the rem…

From what I understand you shouldn't wait more than 5min between prompts without compacting or clearing or you'll pay for reinitializing the cache. With compaction you still pay but it's less input tokens. (Is compaction itself free?)

>pay for reinitializing the cache

Why can't they save the kv cache to disk then later reload it to memory?

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#408

Post by Bridged7756 »

danny_codes wrote: post #251
Bridged7756 wrote: post #233

Earlier quoted context omitted.

Mirrors my sentiment. Those tools seem mostly useful for a Google alternative, scaffolding tedious things, code reviewing, and acting as a fancy search. It seems that they got a grip on the "coding LLM" market and now they're starting to seek actual profit. I predict we'll keep seeing 40%+ more expensive models for a marginal performance gain from now on.

I just don’t see how they’ll be able to make a profit. Open models have the same performance on coding tasks now. The incentives are all wrong. Why pay more for a model that’s no better and also isn’t open? It’s nonsense

I wouldn't say the same but it's pretty close. At this point I'm convinced that they'll continue running the marketing machine and people due to FOMO will keep hopping onto whatever model anthropic releases.

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#409

Post by equasar »

blahblaher wrote: post #56
nate8bit wrote: post #49

Earlier quoted context omitted.

Any recommendations on good open ones? What are you using primarily?

qwen3.5/3.6 (30B) works well,locally, with opencode

The thing I dislike about OpenCode is the lack of capabilities of their editor, also, resource intensive, for some reason on a VM it chuckles each 30 mins, that I need to discard all sessions, commits, etc.

I don't know if it is bun related, but in task manager, is the thing that is almost at the top always on CPU usage, turns out for me, bun is not production ready at all.

Wish Zed editor had something like BigPickle which is free to use without limits.

Re: Anonymous request-token comparisons from Opus 4.6 and Opus 4.7

#410

Post by viktorianer »

hgoel wrote: post #74

The bump from 4.6 to 4.7 is not very noticeable to me in improved capabilities so far, but the faster consumption of limits is very noticeable. I hit my 5 hour limit within 2 hours yesterday, initially I was trying the batched mode for a refactor but cancelled after seeing it take 30% of the limit within 5 minutes. Had to cancel and try a serial approach, consumed less (took ~50 minutes, xhigh effort, ~60% of the rem…

[dead]
Post reply on HN