3 ms·
Open weight SOTA models are not greatly cheaper than Anthropic and providers don't have to cover training capex.
by HeatrayEnjoyer 3mo ago
Open weight SOTA models are not greatly cheaper than Anthropic and providers don't have to cover training capex.
- ElFitz 3mo ago> and providers don't have to cover training capex. How so?
- sethaurus 3mo agoWith an open-weights model, anyone with the hardware to run the model can act as a provider. The median provider pays nothing for the model, so they do not contribute to the cost of training it.
- ElFitz 3mo agoAh! Makes sense. Thanks!
- HeatrayEnjoyer 3mo agoProviders like Fireworks offer inference, they're not training their own models. Yet training is the most expensive part, inference is easy profit.
- ElFitz 3mo agoAnd even the ones who trained it can’t afford to offer it at a price covering their training capex, since any good open-source model will see other providers selling inference without any training expenses to cover. Didn’t see it, but it’s obvious in retrospect. Thanks!