7 ms·
Do you have actual data on the marginal costs of inferencing, or are you using the term 'subsedizing' loosely as in 'discount', meaning a lower price compared t
by PeterStuer 26d ago
Do you have actual data on the marginal costs of inferencing, or are you using the term 'subsedizing' loosely as in 'discount', meaning a lower price compared to another price for the same offer?
- llm_nerd 25d agoLook up the burn rate for OpenAI and Anthropic. Despite the fact that these companies have managed to hook a lot of F500 companies into paying millions for token-level API pricing, they are bleeding cash at a catastrophic rate. And we know subscription tokens come at about a 1/30th the cost of the API pricing (presuming you use your quota). Yes, they are most certainly subsidizing the subscription plans. I mean, at least for people who utilize them to the quota.
- PeterStuer 22d agoThe question is specifically about the marginal cost of the inferencing part exclusively. Not the R&D spend. AFAIK, there has never been a statement on that from neither OpenAI nor Anthropic. What it actually costs to serve one more token determines wether it is "subsidized" at a certain price point. That it is sold for more elsewhere is not evidence.
- llm_nerd 19d agoHow would that number help at all or change the discussion? Is R&D not a real expense? I mean, this argument would have merit if they built something that they're going to monetize for decades, but cutting edge models now grow obsolete in a 6 month time window. There is zero financial analysis where they aren't massively subsidizing the subscriptions, and people have to invent ignorant "oh just ignore most of the cost of providing the product" to try to pretend there is.