I'm curious about what other developers are using these days.
For coding, debugging, code reviews, documentation, architecture discussions, and general development work:
- Do you use OpenAI (ChatGPT/GPT-4o/o3/o1) more?
- Or do you use Claude (Claude 4 Sonnet/Opus) more?
Why?
I'm especially interested in:
1. Which model produces better code?
2. Which model is better at understanding large codebases?
3. Which one do you trust more for production-level work?
4. Has anyone completely switched from OpenAI to Claude (or vice versa)?
Please share your experience and tech stack (.NET, Java, Python, JavaScript, etc.).
Thanks!

Sam HobbsPosted Jun 21, 2026, 9:49 PM
I avoid using Claude but that is a personal preference. I do not use ChatGPT much, especially not for code, but I am not sure why.
I have a free subscription to Perplexity AI and use it often. I sometimes use it for code.
Google Gemini offers a very economical plan and I use it for code.
I sometimes use Microsoft Copilot for code.
The advantage of Github Copilot (not the same as Microsoft Copilot) is that it works in Visual Studio, the full IDE. For generating and fixing code it is very advantageous to use a tool that works in VS. Note that although there are very many Copilots none of them are a model, they orchestrate one or more of a variety of models. I think the main reason I avoid Github Copilot is that it seems more expensive.
If you prefer VS Code then try Cursor, it is a fork of VS Code that has AI built in. I once wrote a specification for a program and used the exact same specification in multiple virtual assistants (AI) and Cursor was the only one that got it right. I asked for a list control that had text items in which the text wrapped and Cursor was the only one that generated code that did that. I sometimes use Cursor to generate a project then use VS for the project after that.
owsiuPosted Jun 21, 2026, 12:53 PM
I’ve been using both. For anyone exploring AI tools beyond ChatGPT and Claude, https://ccnavx.com is worth checking out.