Comment by behnamoh
1 day ago
Anthro is having its Apple moment: too many customers means the company is always on the news, for better or worse.
When iPhones receive negative reviews it's not like only Apple screwed up; others did too, but they sell so much less than Apple that no one hears about them:
"Apple violated my privacy a tiny bit" makes the news;
"Xiaomi sold my fingerprint info to 3rd party vendors" doesn't.
Similarly, Anthropic is under heavy fire recently because frankly, Claude Code is the best coding agent out there, and it's not even close.
You're being downvoted because not only is Claude Code bad at what it does, none of the LLMs can actually code meaningfully, they can just statistically regurgitate other people's code without understanding what it does.
Saying otherwise is just extreme copium.
Anthropic's models are good. Claude Code isn't good.
find codex better by far in some cases
Lmao? seriously. Claude Code is epic levels terrible compared to opencode.
Can you elaborate?
There's been a lot of very bad bugs for a long time and they don't seem to care about fixing them. Eg the screen flicker: https://github.com/anthropics/claude-code/issues/3648
how so?