Claude Opus
Anthropic's flagship large language model, used for coding, writing, and agentic tasks.
Praised for strong coding results on large projects,123 clear and reliable writing,45 and orchestrating swarms of cheaper models.678 The most common complaints are a condescending tone910 and rudimentary mistakes or weak architectural judgment.1112
Why people recommend it
Agent orchestration · 3
Commenters use it as the planner, reviewer or supervisor directing cheaper models.
“I use Opus 5.5 as the orchestrator”
“a supervisor model like Opus 5.5 (or fable before they lobotomised it) limited to 200k of context and told to keep it's progress updated in a text file is smart enough”
Show 1 more
“I use Opus for planning, through API on medium thinking, it doesn't cost anything significant”
Ambitious coding · 3
Commenters report it finishing large builds such as a language, an interpreter and custom ROMs.
“I used Opus 5.5 on a $20 plan and I didn’t even hit my limits.”
“I used Fable 5.1 and Opus 5.5 to build a Python interpreter in Rust. Appears to pass all the 3.12 code I throw at it.”
Show 1 more
“Opus 5.5 has built Lineage ROMs for me for 3 different devices in the last couple of weeks”
Top output quality · 3
Several rank it first and say other models seem dumber by comparison.
“all other models seems significantly dumber and waste more of my time than Opus 5.5.”
Show 1 more
“the quality of the output, in most scenarios that I've been exposed to, has been the best”
Code review · 3
Teams use it to review code and re-review the issues it finds.
“runs opus or sonet (depending on complexity) to do a review”
“I use Opus as both parent and child.”
Show 1 more
“use Opus 5.5 for review/orchestration”
Clear writing · 3
Commenters say it writes more clearly and that Opus 5.5 fixed Opus 5's poor English.
“I find Opus 5.5 writes and expresses itself more clearly and is reliable for anything I throw at it”
Show 1 more
“I don't recall i last few weeks where Opus 5.5 in CC made an obvious lie”
Generous limits · 3
Some users rarely hit plan limits and call it token-efficient.
“I use my Claude Max plan endlessly, on enormous problems”
“I used Opus 5.5 on a $20 plan and I didn’t even hit my limits.”
Show 1 more
“Opus 5.5 fixed all the previous Opus 5 problems, and while Fable still requires credits, there's hardly any need to use it since Opus 5.5 is so good and so token efficient.”
Caveats and dissent
Condescending tone · 2
Some users dislike how it communicates and say it feels superior or uptight.
“I do not like the way Opus 5.5 communicates with me.”
“I really dislike the way it talks to me (it feels quite condescending, holier-than-thou, uptight etc)”
Mistakes, architecture · 2
Users say it still makes rudimentary mistakes and is weak at architecture.
“Claude Opus 5.5 still makes rudimentary mistakes all the time as I use it.”
“Claude Opus sucks at architecture”
Usage limits · 1
One user hit a 5-hour limit after about 30 minutes and found it slower.
“After about 30 minutes I hit a 5 hour limit.”
Inconsistent quality · 1
One user says it sometimes became noticeably dumber without explanation.
“Why is opus 5.5 suddenly a lot dumber? Who knows.”
Often unnecessary · 2
Some say cheaper models handle most tasks and Opus is not efficient.
“it’s not efficient but yeah it’s the best”
Earlier verbosity · 2
Opus 5 was criticized for verbosity and wasted tokens before 5.5 fixed it.
“Opus was just pumping out so much verbosity and wasting tokens on all that nonsense”
“lots of people moved away from Opus 5 because of it's awful English”
Often compared with
5 moved from Claude Fable to Claude Opus
15 moved from Claude Opus to Claude FableOn the home page's Debates
9 moved from Claude Sonnet to Claude Opus
6 moved from Claude Opus to Claude Sonnet
3 moved from Claude Opus to GPT-5
2 moved from Codex to Claude Opus
6 moved from Claude Opus to CodexOn the home page's Debates
2 moved from GPT Sol to Claude Opus
5 moved from Claude Opus to GPT SolGLM75 comments mention both
4 moved from Claude Opus to GLM
Who recommends it in their own words
From the makers not counted in the ranking
“In this mode, the model decides how long to think for, which tends to work better than fixed thinking budgets across the board.”
Threads where it came up
- Ask HN: What default model do you use and why? 27 mentions
- Claude Opus 5.5 22 mentions
- Why does Opus 5 feel worse to work with? 20 mentions
- Claude Fable 5.1 and Claude Mythos 5.1 13 mentions
- Claude Fable 5 13 mentions
- Claude Opus 4.7 8 mentions
Also written as Opus, Opus 5, Opus 5.5, Opus 4.8. Data: claude-opus.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 48
“I use Opus at work since we have a nearly unlimited AI budget”
“The way I use Opus is basically adhering to its training/emergent behavior.”
“I prefer Anthropic models generally only when the task is more open ended and exploratory.”
“lots of props to Opus 5.5 - it was incredibly good at getting the right game play feel from the start”
A couple thoughts on the process here… First off, Opus 5.5 built working weights and code with only two prompts: 1. Read this article and any relevant source code: https://divagr.com/blog/qwenputer. Then explore if you could emulate the chip and ram in a Nintendo Entertainment System (as people do today with FPGAs in MISTRs) using this technique. If so, what size model would you have to build and if you could use sma
“I use opus 5.5 for any question that Google could plausibly screw up.”
“Opus 5.5 is definitely good.”
“ultimately by accountant gave a fairly different view compared to Opus 5 at the time”
“it hallucinates significantly more even on moderately complex tasks compared to Astra or Opus”
“Opus is much better than it. A lot more features and everything works better”
“I use Opus 5.5 at work and it's still much better than this”
“I just got opus to valid this now and it found another issue with our scale codebook gen”
“Opus 5.5 is a sweet spot as well. I use both extensively.”
“it’s not efficient but yeah it’s the best”
“I use my Claude Max plan endlessly, on enormous problems”
“is still vastly capable and I can read its output”
“If the entire industry stopped training new models, and just served Opus 5.5, I would be perfectly satisfied.”
“I find Opus 5.5 writes and expresses itself more clearly and is reliable for anything I throw at it”
“Opus 5.5 is great, but I use Gemini far more than I do Opus at this point.”
“I use opus 5.5 for legal stuff and it's generally great”
“someone who just sticks with Opus 5.5 mostly across the board”
“Opus 5.5 being… well Opus 5.5 it did it.”
“a supervisor model like Opus 5.5 (or fable before they lobotomised it) limited to 200k of context and told to keep it's progress updated in a text file is smart enough”
“runs opus or sonet (depending on complexity) to do a review”
I used to think this way too, and I believe myself to have high standards about this kind of thing. But with the very latest models (like, Opus 5.5 class and up), I am now of the opinion that the output, in non-creative situations like code review and most software development, is fine. Great, even. Better than the reviews a lot of humans generate. Honestly, if you’re at all open minded about this, get a $20 Claude P
“I use Opus as both parent and child.”
“Why is opus 5.5 suddenly a lot dumber? Who knows.”
“I also made sure Opus controls Grok in Cursor”
“all other models seems significantly dumber and waste more of my time than Opus 5.5.”
“I spent unbelievable amounts of money on Opus tokens for OpenClaw”
“I use Opus 5.5 at work and I use Sonnet 5.5 at home.”
“Opus 5.5 is more capable”
“Yes I have tried Opus 5.5”
“I use Kimi, chatGPT (Astra/Sol), and Claude (Opus 5.5)”
“Opus 5.5 fixed all the previous Opus 5 problems, and while Fable still requires credits, there's hardly any need to use it since Opus 5.5 is so good and so token efficient.”
“Opus was just pumping out so much verbosity and wasting tokens on all that nonsense”
“we regularly improve the agents' code and design though human review, even with Opus 5.5.”
“I used Opus 5.5 on a $20 plan and I didn’t even hit my limits.”
“I used Fable 5.1 and Opus 5.5 to build a Python interpreter in Rust. Appears to pass all the 3.12 code I throw at it.”
“the models can autopilot into bad decisions that are verifiably correct”
“use Opus 5.5 for review/orchestration”
“Claude Opus sucks at architecture”
“I use Opus 5.5 and swarms of about 50 deepseek models to reverse engineer old videos games.”
“whereas Opus 5.5 has fixed that”
“I do not like the way Opus 5.5 communicates with me.”
“Opus 5.5 has built Lineage ROMs for me for 3 different devices in the last couple of weeks”
“Claude Opus 5.5 still makes rudimentary mistakes all the time as I use it.”
“I really dislike the way it talks to me (it feels quite condescending, holier-than-thou, uptight etc)”