I've been on Claude Pro (~$20/month) for a while and overall I'm pretty happy with it, but I'm starting to feel the token limits, especially with Claude Code.
I've always had some ethical reservations about switching to OpenAI/Codex – I generally trust Anthropic's approach to safety and alignment more than OpenAI's lately, given everything that's happened. But I'm wondering if I'm being irrational about it.
For context, I'm an indie developer working on personal projects – nothing enterprise-scale, so cost matters.
A few things I'm trying to figure out:
- Is the token burn on Claude Pro a known pain point, or am I just hitting edge cases?
- Has anyone switched from Claude to Codex and actually regretted it (or not)?
- Are there smarter ways to stretch the Pro plan before jumping ship?
Not looking for a flame war, just genuine takes from people who've been in the same spot.