Claude · 近 7 天
Claude Opus 5.5
0–100 分，越高代表社区使用体验越正面，不代表能力测试成绩。
更新于 2026-10-01 09:47 UTC · 2026-09-24 – 2026-10-01 UTC体感分81.2/100972 条评论 · 近 7 天
大家怎么看
有用户认为 Claude 的回复常常过于密集、难以理解，但 Opus 5.5 在给出更自然、更像正常人的回复方面略有改进。有用户反映 Opus 5.5 会拒绝执行请求，但有一次用户解释说是在做游戏，模型就配合完成了。
样本与评分说明
截至 2026-10-01 09:52:04 UTC，Claude 家族的 Claude Opus 5.5 在近 7 天有 972 条明确归属该版本的社区反馈，社区口碑分为 81.2/100。部分有效分类评分：通用文本 78.6/100 (n=363)；编程 84.2/100 (n=105)；推理 89.0/100 (n=287)。 评分方法与数据来源
社区评论
精选近 7 天的不同意见，摘录条数不代表真实好评比例。
38 条精选摘录
额度评价来自社区体验，不能据此推算官方实际配额。
展开原文
I find fascinating the complexity on some parts tho.
展开原文
Opus 5.5 is a little better at normal responses though lol
查看评论上下文
虽然我理解它在说什么，但很多时候它表达的方式太密集了，一个人本来可以压缩/简化这个意思，这样我就不必花光所有脑力去理解这个句子，而是直接理解逻辑/问题哈哈哈。
机器翻译展开原文
While I understand what it’s saying, a lot of the times it says things in such a dense way where a human could condense / simplify the meaning so i don’t have to spend all my brainpower understanding the sentence and actually get to the logic/problem lmfao.
展开原文
5.5 speaks actual English
展开原文
also set to just use opus 5.5 for harder tasks
展开原文
I’ve had opus5.5 push back… in one case I explained it’s a game, and yeah, we made a game that does exactly what I wanted to do.
展开原文
I highly doubt any amount of tinkering with your qwen 3.8 27b setup would deliver the quality Opus 5.5 can do
展开原文
in terms of creativity, creating new things, yeah opus 5.5 is fantastic
展开原文
Astra and Opus 5.5 are great, but even what they make might not even be worthy the lowest tier indie games.
展开原文
I can’t even put reasoning traces in front of opus 5.5 without hitting a “reasoning extraction” claude code notice guardrail?
展开原文
it catches many mistakes made by Opus 5.5
查看评论上下文
SOL 5.6 和 6.1 都是高水平后端编码的怪兽。
机器翻译展开原文
SOL 5.6 and 6.1 both are monsters for high level backend coding.
展开原文
Opus 5.0 and 5.5 (still) feel more like steering dependant
展开原文
i am still getting the same push back
查看评论上下文
我刚尝试切换到"接受编辑开启"模式，我仍然遇到同样的拒绝。
机器翻译展开原文
I just tried switching to the "accept edits on" mode, i am still getting the same push back.
我会手动登录我的网上银行，我会让 Claude 选择收款人，然后通过 playwright 或 chrome connect 输入支付金额。我将是授权支付的人，但使用 Opus 5.5 和我尝试过的其他几个版本（包括 Sonnet），它似乎拒绝做任何工作！
机器翻译展开原文
I would manually log into my online banking, I would let Claude select the payees and then insert the amount to pay via playwright or chrome connect. I would be the one authorising the payment, but with the Opus 5.5 And a few other versions that I've tried, including Sonnet it seems to refuse to do any of the work!
展开原文
made a very simple mistake, and another simple mistake, again
展开原文
nearly unlimited use
展开原文
it is also little slow even compared to Astra
展开原文
It solved all the issues I had in 2 days, when I was thinking about working for 10 days with GPT
展开原文
I thought Astra was good until I had Opus 5.5 look at the code it generated
展开原文
Opus 5.5 is better at code reviewing than 6.1 Sol
展开原文
holy fuck he is getting me crazy on how slow he is at doing tasks
展开原文
waiting on my weekly reset there
查看评论上下文
Opus 5.5 迄今为止要好得多。
机器翻译展开原文
Opus 5.5 is by far much better.
不过，已经用完了我每周（5倍计划）用量的 76%。
机器翻译展开原文
Used up 76% of my weekly (5x plan) usage, though.
展开原文
Agent rules are more strongly adhered to than claude.mds are at least with Opus 5.5
展开原文
outright ignores what the user has stated as facts and doesn't ask when those dont align
展开原文
Opus 5.5 is a mind blowing leap and barely sips use.
展开原文
I get about the same amount of usage with opus 5.5 on the claude 20$ sub as I get with astra on the 200$ plan. very worth switching.
展开原文
claude opus 5.5 in VS code extension does not
查看评论上下文
截至最新更新，我甚至无法让 Opus 5.5 在我的测试数据库中创建一个虚构用户来测试登录界面，因为它完全拒绝处理密码输入。
机器翻译展开原文
As of the latest update, I cannot even get Opus 5.5 to create a fictional user in my test database to test the login UI because it refuses to handle password inputs altogether.
展开原文
Sure, Opus 5.5 benchmarks better than Fable. Sure. But is that the model, or is that the RL for agentic work?
展开原文
Opus 5.5 (I use it for work writing C#) seems faster than Opus 5, but it doesn't seem demonstrably better to my eyes and is still prone to word vomit
展开原文
I can run Claude Opus 5.5 on x high for 12 hours and burn 12% of my week
展开原文
Opus 5.5 + Fable 5.1 on a $200 plan/heavy workload/single task eats about 15%/day
展开原文
I have yet to have Opus 5.5 block me from doing any of my usual security testing, validation, and remediation work under the CVP.
展开原文
Opus is clearly better in logic and processing steps, but often skips some or does something else entirely
展开原文
graphics would be 10x better
查看评论上下文
它始于我输入到Claude Code中的一条提示词：“*仅使用像素直接重制一个精美的宝可梦红重制版……绝对的像素艺术完美。
机器翻译展开原文
It started from one prompt I typed into Claude Code: "*using only pixels directly remake a beautiful pokemon red remake … absolute pixel art perfection.
* 完全没有图像文件。 游戏逐像素绘制到 320×180 的屏幕上。
机器翻译展开原文
* No image files at all. The game draws into a 320×180 screen pixel by pixel.
展开原文
Opus 5.5 has significantly better vision, which might mean for game dev you're using it more
展开原文
Fr I cant stress enough how amazing Opus 5.5 is in that category.
查看评论上下文
我把我的 Minecraft 模组文件给了 Opus 5.5，并告诉它制作一个预告片
机器翻译展开原文
I gave Opus 5.5 my Minecraft mod file and told it to create a Trailer
展开原文
Its super fast too. Not better than astra but atleast 5x faster
展开原文
Opus 5.5 provides the smoothest LLM experience i ever had, but it did one very odd thing where my macbook ram went full and i had to restart the machine to continue
展开原文
Neither Astra nor Opus 5.5 could recognize that a UI design with cropped borders—caused by insufficient padding—was problematic.
近 7 天暂时没有符合此筛选条件的精选评论。