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

Claude Opus 4.7

anthropic.com

Post reply on HN
Print view View original

881–890 of 1001 posts

Re: Claude Opus 4.7

#881

Post by cageface »

onlyrealcuzzo wrote: post #195
buildbot wrote: post #12

Too late, personally after how bad 4.6 was the past week I was pushed to codex, which seems to mostly work at the same level from day to day. Just last night I was trying to get 4.6 to lookup how to do some simple tensor parallel work, and the agent used 0 web fetches and just hallucinated 17K very wrong tokens. Then the main agent decided to pretend to implement tp, and just copied the entire model to each node...

I switched to Codex and found it extremely inferior for my use case. It is much faster, but faster worse code is a step in the wrong direction. You're just rapidly accumulating bugs and tech debt, rather than more slowly moving in the correct direction. I'm a big fan of Gemini in general, but at least in my experience Gemini Cli is VERY FAR behind either Codex or CC. It's both slower than CC, MUCH slower than Codex,…

I've had exactly the opposite experience. Getting great results using GPT for hours every day since 5.3. You need to put the effort level on at least high though.

Every time I hand off a task to Opus to see if it's gotten better I'm disappointed. At least 4.7 seems to have realized I have skill files again though.

Re: Claude Opus 4.7

#882

Post by berkes »

berkes wrote: post #864
paulluuk wrote: post #836

Earlier quoted context omitted.

If open weight models are sufficient for your engineering problems, then you should absolutely use them. But I haven't seen a single open weight model that can get even close to the complexity in my projects. They sometimes work for small toy examples or leetcode puzzles, but not very any real project. Really curious what models you've found that could replace current state of the art.

I've been using devstral2 with great success for a few months now. The hosted version, not running one locally or such. Devstral is open. Devstral is good, Opus better. But not much. For me, "good" is "good enough". The difference, IME lies in context engineering: skills, agents.md, subagents, tools, prompts. A Devstral with good skills performs far better than an "blank" claude code. Claude with good skills performs…

Someone just asked my what I dislike most about Mistral and about Claude code.

I run both in zed editor. Claude codes' integration is subpar - it's ACP does not report tasks, doesn't give diffs and so on.

Mistral has rate limits that I hit just too often. I'm now using Mistral Pro, where this is worse, using pay-as-you-go is better but costs me 10x the pro. The agent then stops with an error.

Re: Claude Opus 4.7

#883

Post by adammarples »

johnmlussier wrote: post #242
johnmlussier wrote: post #227

They've increased their cybersecurity usage filters to the point that Opus 4.7 refuses to work on any valid work, even after web fetching the program guidelines itself and acknowledging "This is authorized research under the [Redacted] Bounty program, so the findings here are defensive research outputs, not malware. I'll analyze and draft, not weaponize anything beyond what's needed to prove the bug to [Redacted]. I…

⎿ API Error: Claude Code is unable to respond to this request, which appears to violate our Usage Policy (https://www.anthropic.com/legal/aup). This request triggered restrictions on violative cyber content and was blocked under Anthropic's Usage Policy. To request an adjustment pursuant to our Cyber Verification Program based on how you use Claude, fill out https://claude.com/form/cyber-use-case?token=[REDACTED] Ple…

Don't forget that you can also write code by hand

Re: Claude Opus 4.7

#885

Post by morgengold »

Tried it for different Vue, Nuxt, Supabase projects. Think of CRM SAAS or Sales App like size. Also for my personal bot with which i communicate via telegram.

First feelings: Solves more of the complex tasks without errors, thinks a bit more before acting, less errors, doesnt lose the plot as fast as 4.6. All in all for me a step further. Not quite as big of a jump like 4.5 -> 4.6 but feels more subtle. Maybe just an effect of better tool management. (I am on MAX plan, using mostly 4.7 medium effort).

Re: Claude Opus 4.7

#886

Post by adammarples »

driverdan wrote: post #750
ACCount37 wrote: post #352

Earlier quoted context omitted.

This is the same paranoid, anxious behavior that ChatGPT has. One hell of a bad sign.

Models are not paranoid or anxious, they do not think or have feelings. I know you're probably using those words as a metaphor but we need to be careful about anthropomorphizing LLMs.

They didn't describe the model, they described (accurately) the behaviour. They are useful descriptors of behaviour.

Re: Claude Opus 4.7

#888

Post by Traubenfuchs »

Anthropic‘s throwing out new models but the devs are NOT happy.

Was all the goodwill people had for Anthropic products them selling unsustainably high performance at a loss?

Re: Claude Opus 4.7

#889

Post by VA1337 »

Guys, this may have already sounded, but there is a strong feeling that before the release of a new model, they are numbing the previous one

Re: Claude Opus 4.7

#890

Post by krzat »

LostMyLogin wrote: post #431
avaer wrote: post #297

Earlier quoted context omitted.

> Still trying to get a decent pelican out of this one but the new thinking stuff is tripping me up Wouldn't that be p-hacking where p stands for pelican?

Input: I want to wash my car. The car wash is 50 meters away. Should I walk or drive? Output: Walk. It'll take you under a minute, and driving 50 meters barely gets the engine warm — plus you'd just have to park again at the other end. Honestly, by the time you started the car, you'd already be there on foot. --- I asked it to figure out why it made the mistake: "Physical/spatial common sense. Exactly what just happe…

Humans tend to confabulate when asked "why you did X", funny how LLMs are pretty much the same.
Post reply on HN