TodayToday / Industry / Story180B Parameters, 32GB of RAM: Pushing MoE Inference to the Limit with TransformersShareWed 18 Nov · 12:00 AM, updated 12:00 AMFirst seen Wed 18 Nov · 12:00 AMUpdated 12:00 AMLimited coverage, taperingHugging Face · IndustryReported by Hugging Face1 source ↓SourcesHugging Face12:00 AMRead ↗180B Parameters, 32GB of RAM: Pushing MoE Inference to the Limit with TransformersImagesHugging FaceHugging FaceImage via Hugging Face(opens the article in a new tab)More on Hugging FaceUS antitrust regulators are scrutinizing Nvidia's Hugging Face deal and sales practicesTue 27 Oct · 10 outletsThinking Machines Lab releases Inkling 2 Mini open modelFri 20 Nov · 2 outlets