Did they get rid of the option to clear the context and work just with the plan, in plan mode? I always used that and it worked well. Now it seems to be gone.
Claude Opus 4.7
861–870 of 1001 posts
Re: Claude Opus 4.7
#862Is there a classic web interface? (noscript/basic (x)html)
Re: Claude Opus 4.7
#863I'm finding the "adaptive thinking" thing very confusing, especially having written code against the previous thinking budget / thinking effort / etc modes: https://platform.claude.com/docs/en/build-with-claude/adapti... Also notable: 4.7 now defaults to NOT including a human-readable reasoning token summary in the output, you have to add "display": "summarized" to get that: https://platform.claude.com/docs/en/build-…
The reasoning modes are really weird with 4.7 In my tests, asking for "none" reasoning resulted in higher costs than asking for "medium" reasoning... Also, "medium" reasoning only had 1/10 of the reasoning tokens 4.6 used to have.
Claude is known for its shitty metering.
Re: Claude Opus 4.7
#864I can't notice any difference to 4.6 from 3 weeks ago, except that this model burns way more tokens, and produces much longer plans. To me it seem like this model is just the same as 4.6 but with a bigger token budget on all effort levels. I guess this is one way how Anthropic plans to make their business profitable. During the past weeks of lobotomized opus, I tried a few different open weight models side by side wi…
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 even better, but hardly noticable, IME.
I am convinced I've plateaued. Better performance comes from improving skills and other "memory", prompting smarter, better context management and, above all, from the tooling around it and the stability of the services.
I do still run Claude with Opus alongside Mistral with Devstral2. Sometimes to just compare outputs, often to doublecheck, but mostly to doublecheck my statement that the difference between Devstral2 and Opus is marginally and easily covered by better context engineering.
Re: Claude Opus 4.7
#865[dead]
Re: Claude Opus 4.7
#866Earlier quoted context omitted.
I don't know, I think java is the best programming language. I use it for everything I do, no other programming language comes close. Python lost all my trust with how slow it's interpreter is, you can't use it for anything. ^^^^ Sarcastic response, but engineers have always loved their holy wars, LLM flavor is no different.
Java is great and all but if you don't use it with the right kind of keyboard you're wasting your time. I use one of those very loud clacky ones with brightly colored keys and that makes me a better person
Joke's on you, I use Java as Clojure with a clacky split keyboard, feels great.
Re: Claude Opus 4.7
#867Did they get rid of the option to clear the context and work just with the plan, in plan mode? I always used that and it worked well. Now it seems to be gone.
It just repopulates the context. It's absolutely infuriating the way it behaves now, since there are not many workarounds to minimize token usage unless you use caveman [1].
Re: Claude Opus 4.7
#868If Claude AI is so good at coding, why can't Anthropic use it to improve Claude's uptime and fix the constant token quota issues?
Maybe this is the result
Re: Claude Opus 4.7
#869It is capable of particularly beautiful writing.
I've had a really nice user preference for writing style going. That user preference clicks better into place with 4.7; the underlying rhythm and cadence is also mich more refined. Rhythm and cadence both abstract and concrete – what is lead into view and how as well as the words and structures by which this is done. The combination is really quite something.
Re: Claude Opus 4.7
#870Opus 4.7 is a slight regression over 4.6 https://petergpt.github.io/bullshit-benchmark/viewer/index.v...
Max is worse than High.