Get the latest tech news
$2 H100s: How the GPU Rental Bubble Burst
H100s used to be $8/hr if you could get them. Now there's 7 different places sometimes selling them under $2. What happened?
Eugene has now cofounded Featherless.AI, an inference platform with the world’s largest collection of open source models (~2,000) instantly accessible via a single API for a flat rate($10-$75+ a month). While any layer down the stack may be vertically integrated (skipping the infra players for example), the key drivers here are the “Resellers of unused capacity” and the rise of “good enough” open weights models like Llama 3, as they are all major influencing factors in the current H100 economical pressures. At Featherless.AI - We currently host the world’s largest collection of OpenSource AI models, instantly accessible, serverlessly, with unlimited requests from $10 a month, at a fixed price.
Or read this on Hacker News