alt Hacker NewsSerous question - why do people stick with Clause Code over Cursor? With Cursors base subscription I have access to pretty much all the Frontier models and can pick and choose. Anthropic models haven’t been my go-to in months, Gemini and Codex produce much better results for me.
Replies
My answer is that I tested both, and Claude Code (~8 months ago) was so obviously better than Cursor that I continue to happily pay Anthropic $200/month. Based on anecdotes I happen to catch, I don't believe Cursor's caught up.
The value isn't just the models. Claude Code is notably better than (for example) OpenCode, even when using the same models. The plug-in system is also excellent, allowing me to build things like https://charleswiltgen.github.io/Axiom/ that everyone can benefit from.
Because when it's good, it's really good - Cursor doesn't work as well for me and also I prefer the TUI experience. If anything, the real alternative is OpenCode.
Part of the sauce is not in the model, but in the agent itself. And for that matter, I think AMP an incredibly better agent that Claude Code. But then, Claude heavily subsidized subscription prices are hard to beat.
Wouldn't you run out of tokens sooner? That's the big problem.
Because I tried all the Cs - Copilot, Cursor, Codex, and Claude - and Claude consistently have better results. Codex was faster, Copilot had better integration, Cursor sometimes seemed smarter, but Claude was the best most reliable consistent experience overall, so Claude is what I stuck with - and so did the rest of our eng department.
Cursor performs notably worse for me on my medium-sized codebase (~500kloc), possibly because they try to aggressively conserve context. This is especially true for debugging, Claude Code will read dozens of files and do a surprisingly good job of finding complex bugs, while Cursor seems to just respond with the first hypothesis it comes up with.
That said, Cursor Composer is a lot faster and really nice for some tasks that don't require lots of context.