Is it worth switching from GPT to Claude?
retained text
Hey everyone, I’ve been using both GPT and Claude models, but only through the API. Over time I’ve leaned more toward GPT because of the price-to-quality ratio, and I’m now thinking about getting a Claude subscription since OpenAI is reducing usage limits. One thing I like about GPT is that it tends to create test files (unit tests or other kinds of checks) to verify that the code it writes is correct. In the little testing I did with Claude models, they didn’t do this by default. So my questions are: 1. Have you noticed this difference between GPT and Claude? Is Claude less likely to write tests on its own unless you ask? 2. When you use Claude (in the app, Claude Code, or the API), do you add a skill, a system prompt, or a CLAUDE.md-style instruction that makes it write and run tests? Could you share what works for you? 3. For those who switched from GPT to Claude, how does the coding quality compare, and is the subscription worth it compared to paying per token? Any tips or experiences are appreciated. Thanks!
document fetched 2026-10-30 00:21 · canonical https://reddit.com/r/ClaudeAI/comments/1xiudqe/is_it_worth_switching_from_gpt_to_claude