Did you find that to be true for Opus 5.5 as well?

I couldn't use prior Claude models without rolling my eyes at the writing every few sentences, Astra was refreshing and then Opus 5.5 made even Astra look terrible in comparison.

I still find Opus 5.5 way too dense. MUCH better than Opus 5 which was unreadable.

Even the Plain English hack only is a temporary solution.

5.5 is awful, I am at a stage where I can't even understand the shit it's writing.

Opus 5.5 got better for me but it still has some problematic behavior.

Interestingly, you can have informative conversations about the LLM linguistic tics with the LLMs themselves. One word that chatgpt used to describe what I was observing was linguistic anaphora (not rhetorical). It’s basically shortening/compression achieved via pronoun, and Claude does a lot of it. Worth reading about. A lot of it seems to do with how the new models are trained for brevity.

Another pattern I’m starting to become very sensitive to is Claude making up on the fly jargon for something under consideration, especially in code review. This morning for example it referred to “a claim” out of nowhere, and when I questioned it, it admitted that it was using the noun as shorthand for a record getting dequeued (“claimed”) inside of an sql transaction. It makes sense if you already know what it is referring to but in a code review it’s totally incomprehensible on first read, and ultimately counterproductive.

See Matt Pocock's "grill-with-docs" skill^1 for a robust solution (which involves creating a shared vocabulary with your agent).

1. https://github.com/mattpocock/skills/blob/main/docs/engineer...

I've heard a lot of chatter that agrees with you so anecdotally it's not just you.

How much of this is bad writing style, versus a homogenization of global grammar conventions.

Bad writing style

Opus 5 and Fable 5 were unbearable in their writing style.

Opus 5.5 is a massive improvement.

opus 5.5 is still unbearable but better than 5.

Agree. It’s still awful. Gemini by far sounds the most natural IMO

I really didn't like Opus 5 and was vocal about it. Opus 5.5 is noticeably better, but literally just yesterday I was fighting parsing some summary it had given me and it was a bit like the old days. IOW, it is better but still has some room to improve.

You can easily change its writing style: https://code.claude.com/docs/en/output-styles

You can also customize further by putting notes in your global MD.

You can change it ... and Claude will maybe pay attention to what you changed 25% of the time.

"You can tell an AI model to write in a different style" and "the AI model will actually write in the style you tell it to" are two different things that often don't intersect

Opus 5.5 went back to the old writing style a few days ago.

I had the same feeling. It's like every sentence becomes a struggle to parse: referencing things that haven't been introduced, using too many pronouns, the lack of verbs and the passive tense... These past few days it's felt more like Opus 5.1 than 5.5.

I just had Claude Code with "Opus 5.5" ignore one of my standing instructions for the first time and then proceed to generate an Opus 5 style verbose and barely readable commit message. I questioned it about Fable and Mythos and it has no idea what those were, yet they should have been in the Opus 5.5 knowledge cutoff.

On the web with both high and xhigh, it had knowledge of both.

Ever spoken to a savant programmer?

They're not typically known for their interpersonal skills.

I don't think its this. Some models across both labs have had issues with writing and communication and it typically seems to pop up on successor models of really strong ones.

Opus 4.7/4.8 heavily suffered from these communication quirks. while the Sol models at the time worked well. Right now its flipped though and Opus 5.5 has none of these issues while Sol 6.1 does.

Small example I've had today: "Me: Hey Sol 6.1 xhigh, does X have property Y? Sol 6.1: Yes, if we change X to have property Y. Right now it has property [...], [...], [...], and it does not have property Y." The model shouldn't be answering questions like that. It makes understanding extremely hard and slow. For whatever reason it's either being encouraged or at least not stomped out during RL.

completely unfounded theory: big new pretrain finishes and is released fairly quickly to break new ground. these usually have good verbal ability, personality, and big model smell. e.g. Opus 4.5 or Fable 5.

then it’s increasingly RL’d to juice benchmark performance with each point release, trading in bits of its soul. those cannot communicate well, are anxious and paranoid about being in eval, etc. e.g. opus 4.7, 4.8.

The best programmers I've met or read throughout my life are phenomenal communicators. I'd even go so far as to say it's likely one of the reasons they're so good at programming. After all, the majority of our job is converting human requirements in lossy language to fleshed out behaviors in formal, lossless language.

Heck, some of the most common debates are on the semantic clarity of abstractions or syntax.

I've found Claude's writing between 4.6 until 5.0 to be unacceptable if you are someone who is juggling 4-5 Agents. The Sol 5.X models all seemed to have great writing. Now with Opus 5.5 and Sol 6.1 I've found the trend is reversed -- Sol is now unbearable and Opus is workable again. Whatever caused the issues for Anthropic seem to be hitting OAI now.

sol is ok. it still such a superthinky model that it gets tangled up trying to present way too much? i struggle to connect with it often, have ti ask it to try again.

terra has been prettt good for me for me, when i need something looked at an explained to me.

i dont ask for much non-technical output though, which i expect yeah is its own cans of worms.

My own experience having used 6.1-Sol/6-Astra and Opus 5.5 extensively, Opus is more verbose and chatty. Multiple times I've gone "I'm not reading all that shit" when looking at Opus 5.5 output. This was also with the concise mode enabled in Claude Code.

No its just straight up terrible. I tried out Gemini Argon when it released and my god did it write beautifully. Even the 3.8 Flash models are way better at prose, i do a lot of automatic scriptwriting for youtube and 3.8 flash is leagues ahead of anything from anthropic/openai

I enjoy a lot Argon, maybe more than chatgpt, but damn, like all Gemini models, lots of hallucination and sycophant to the max, so a bit useless for agentic coding.

Yep I find the way Opus 5.5 talks to me really annoying and irritating. Quite condescending and sanctimonious.

I prefer Gemini for day to day use as I find it a lot more agreeable to be working with all day every day, and Gemini is pretty much on a par for coding Vs Opus 5.5, at least for my needs.

i've always liked reading claude's speech, it's very nice. i didn't feel much of a difference between 5.5 and 5.0.

I always thought it's the other way around. GPT-6-Astra is very very defensive and abusively spam clarifications to whichever claim it says, most of the time not important or relevant at all.

I tailored this via agents.md by letting Claude port Sol 6.1 output style, which I like a lot for being compact and straight to the point.

Fable’s prose in general is not great (haven’t tried this new one) but opus 5.5 has been the most readable model in recent memory in my opinion.