Dmytro Dzhulgakov(@dzhulgakov):periodic reminder: cache hit rate and model token efficiency (# of turns, # reasoning tokens) matter more for cost than per-token sticker price working on more updates to ⬆️ cache hit rate on @FireworksAI

Dmytro Dzhulgakov

@dzhulgakov

Co-founder and CTO @FireworksAI_HQ. PyTorch core maintainer. Previously FB Ads. Ex-Pro Competitive Programmer

加入 April 2013

829 正在關注 7.7K 粉絲

Dmytro Dzhulgakov@dzhulgakov

2026.08.31 17:42

periodic reminder: cache hit rate and model token efficiency (# of turns, # reasoning tokens) matter more for cost than per-token sticker price working on more updates to ⬆️ cache hit rate on @FireworksAI_HQ

Thibault Jaigu@ThibaultJaigu

2026.08.30 17:59

same model. same list price. 5.7x apart in real cost. glm-5.2 across 6 hosts: @FireworksAI_HQ 18% of list @sferenceai 30% @tensorx_ai 43% @nebiustf 100%, caches nothing the price sheet tells you almost nothing. cache hit rate is the price.