[timpera]

In my experience, OpenAI gives you unreasonable amounts of compute for €20/month. I am subscribed to both and Claude's limits are so tiny compared to ChatGPT's that it often feels like a rip-off.

Claude also doesn't let you use a worse model after you reach your usage limits, which is a bit hard to swallow when you're paying for the service.

[replwoacause]

Same experience here. I started our devoutly using Claude but ran into some many limits that I switched back to ChatGPT and it's been night and day. I haven't even really been able to play with the Opus model on my Pro plan because it devours usage and then blocks me for X hours until it resets, costing me a work day. OpenAI has never done that to me. In fact, Codex just churned away for 2 hours on a task and I'm still using it without hitting a limit. I used to love using Claude but the limits are too prohibitive.

[appsoftware]

Claude when used via Github Co-Pilot is much better for useage allowance. I used Opus 4.5 for a months worth of development and only just hit 90 pct of the pro $40 per month allowance.

[lm28469]

If their pay as you go api token prices reflect their internal costs then it makes sense, but it could also be that claude makes money while gpt sells at loss to stay on top. Claude is way more expensive overall, and way more limited with flat rate subscriptions

opus: 5/25 gpt: 1.75/14

[int_19h]

Given how much you can use Codex on their $200 plan, I'm virtually certain that it's subsidized.

As to why, I think in part it is because people who are willing to pay that much per month are much more likely to be using it heavily on "serious" tasks, which is, of course, a goldmine for training data - even if you can't use the inputs directly for training, just looking at various real world issues and how agents handle them (or not) is valuable, especially when all the low-hanging fruit have already been picked.

I wouldn't even be surprised if the $20 users are actually subsidizing the $200 users.