And the real numbers could be better for Anthropic. It's feasible Opus models are actually cheaper to serve than GLM 5.2 because Anthropic have optimized the hell out of inference.
I guess that’s the big question, will people pay a big margin long term to use their end products / models or will AI tokens be commoditized by many competing players. For coding if I had to pay API costs I’d switch in a heartbeat, enterprise maybe more reluctant?
I know some down to about 1% ($200 Max plan vs $20k in tokens per month)
Not every subscriber is a full time SWE. In fact most professional SWEs will be on enterprise plans and thus not get subscriptions at all.
I think it's very plausible that subscriptions are overall losing money. But we simply don't know.
(and if you are using per-token billing as an enterprise user without first maxing out a premium seat you are very silly)
anthropic probably has more customers that use more of their sub, but for open ai where a lot of their subs are consumers through chatgpt.com, they have a lot of free money to work with there