> it states that OpenAI choose not to disaggregate prefill and decode
They disaggregate INSIDE the chip, not by having separate machines for the 3 phases. the slides:
https://x.com/beffjezos/status/2092416851737518190
> it states that OpenAI choose not to disaggregate prefill and decode
They disaggregate INSIDE the chip, not by having separate machines for the 3 phases. the slides:
https://x.com/beffjezos/status/2092416851737518190