Cancelled Claude Code Pro
Posted on: 2026-03-12
At the beginning of the year, I subscribed to a year of Claude Code Pro (200$). Last weekend, I cancelled it. The reason is simple: I was hitting the limit every evening after 30 to 60 minutes. Instead of upgrading from Pro to Max, which would have increased the price by 10x, I decided to try OpenAI Codex for 20$. My reasoning was that if I could use about 45 minutes of Claude and 45 minutes of Codex each day, that would probably be enough for my needs.
To my surprise, I’ve ended up using Codex every day. Last weekend I even used it for 5 hours straight and never hit a limit. For 20$, I suddenly had full access. I was never stopped in the middle of a prompt or asked to pay extra. I’ve been trying it for over two weeks now, and most days I use it between 1 and 3 hours without any issue. At this point, I’m barely opening Claude anymore.
This experience made me realize that the model itself isn’t really the issue. Opus 4.6 and GPT-5.4 feel mostly similar in practice. The real difference is the experience around them. OpenAI removes a lot of the friction. Once you pay, you can simply use the tool and experiment without worrying about hitting a limit.
I’m aware this pricing dynamic may not last forever. It reminds me of the early days of Netflix, when it was about 8$ and had an enormous library. That obviously changed over time. But for now, it’s financially more attractive for me to pay 20$ to OpenAI than to continue paying for Anthropic’s higher tiers.