Greetings everyone,
I am Thomas, a computational biophysicist / cheminformatician, currently upskilling in GenAI. I hope it is allowed to ask the following question in this mailing list.
My ChatGPT Business subscription is ending, and I’m considering switching to a personal plan. Which plan do you use for coding and research—ChatGPT Plus ($20/month), ChatGPT Pro ($100/month), Claude Pro ($20/month or $200/year), or Claude Max 5x ($100/month)?
Are the usage limits sufficient for your monthly coding, web search, and research work, or do you regularly hit the limits? I’d appreciate your recommendations, especially regarding Codex versus Claude Code.
Allegedly, the pricing of GPT-5.6 Luna is beyond match.
Hey Thomas,
I’ve used both ChatGPT Plus and Claude Pro for coding work, and here is how I see they break down in practice:
Claude Pro (Claude Code)
- Usage Limits: In Claude Code, with a ~200k context window, active coding sessions (~44k tokens input/output) typically hit the rate limit after 3–5 hours of heavy use, requiring a 5-hour wait for the quota to reset.
ChatGPT Plus (Codex)
- Usage Limits: ChatGPT Plus offers around 160 messages every 3 hours for flagship models. Once hit, it gracefully falls back to a faster, lighter model rather than hard-blocking you.
If your workflow relies on broad repository awareness and long context reasoning, Claude Pro (or upgrading to Claude Max if you hit limits daily) is hard to beat. But if you prefer continuous uptime without hard lockouts and strong standalone generation/sandboxing, ChatGPT Plus (or ChatGPT Pro for unrestricted reasoning access) is very reliable. Also some developer prefer ChatGPT Plus over Claude as they believe Codex produces a “cleaner” and optimized code.
Hope it helps you.
