Opus 4.7 came even quicker than I expected. It's like they are releasing a new Opus to distract us from Mythos that we all really want.
Claude Opus 4.7
561–570 of 1001 posts
Re: Claude Opus 4.7
#562We've all been complaining about Opus 4.6 for weeks and now there's a new model. Did they intentionally gimp 4.6 so they can advertise how much better 4.7 is?
Re: Claude Opus 4.7
#563Earlier 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…
I tried o3, instant-5.3, Opus 3, and haiku 4.5, and couldn't get them to give bad answers to the couch: stairs vs elevator question. Is there a specific wording you used?
Re: Claude Opus 4.7
#564This is a CC harness thing than a model thing but the "new" thinking messages ('hmm...', 'this one needs a moment...') are extraordinarily irritating. They're both entirely uninformative and strictly worse than a spinner. On my workflows CC often spends up to an hour thinking (which is fine if the result is good) and seeing these messages does not build confidence.
Sounds really minor, but was actually a big contributor to me canceling and switching. The VS Code extension has a morphing spinner thing that rapidly switches between these little catch phrases. It drives me crazy, and I end up covering it up with my right click menu so I can read the actual thinking tokens without that attention vampire distracting me. And of course they recently turned off all third party harness…
I used Gemini CLI for a while because it was free to me. The primary reason I stopped was because it wasn't very good, but their "thinking summaries" didn't help matters. They were model generated and just said things to the effect of "I'm thinking very hard about how to solve this problem" and "I'm laser-focused on the user objective". So I feel you: small things like this make a big difference to usability.
Re: Claude Opus 4.7
#565Working on some research projects to test Opus 4.7. The first thing I notice is that it never dives straight into research after the first prompt. It insists on asking follow-up questions. "I'd love to dive into researching this for you. Before I start..." The questions are usually silly, like, "What's your angle on this analysis?" It asks some form of this question as the first follow-up every time. The second obser…
[dead]
Re: Claude Opus 4.7
#566The benchmarks of Opus 4.6 they compare to MUST be retaken the day of the new model release. If it was nerfed we need to know how much.
https://marginlab.ai/trackers/claude-code-historical-perform...
Surely they are testing their optimizations against common benchmarks internally? I bet the "real world task" degradation is larger by some multiple than it appears when measured through a benchmark that is part of the target.
Re: Claude Opus 4.7
#567I'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-…
https://github.com/anthropics/claude-agent-sdk-python/pull/8... - created PR for that cause hit it in their python sdk
Re: Claude Opus 4.7
#568I'm running it for the first time and this is what the thinking looks like. Opus seems highly concerned about whether or not I'm asking it to develop malware. > This is _, not malware. Continuing the brainstorming process. > Not malware — standard _ code. Continuing exploration. > Not malware. Let me check front-end components for _. > Not malware. Checking validation code and _. > Not malware. > Not malware.
Just happened to me and I was really confused. First time I've seen any malware callouts so it had me worried for a minute.
> This file is clearly not malware
Yeah, it's all my code, that you've seen before...
Re: Claude Opus 4.7
#569I'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-…
... here's the pelican, I think Qwen3.6-35B-A3B running locally did a better job! https://simonwillison.net/2026/Apr/16/qwen-beats-opus/
You used a secret backup test! Truly honored to see the flamingos. We obviously need them all now ;-)
Re: Claude Opus 4.7
#570Earlier quoted context omitted.
Its especially concerning / frustrating because boris’s reply to my bug report on opus being dumber was “we think adaptive thinking isnt working” and then thats the last I heard of it: https://news.ycombinator.com/item?id=47668520 Now disabling adaptive thinking plus increasing effort seem to be what has gotten me back to baseline performance but “our internal evals look good“ is not good enough right now for what ma…
you're using a proprietary blackbox
paying for - so some form of return is expected.