Difference between revisions of "Blackwell (microarchitecture)"

Latest revision as of 04:07, 6 May 2024

Blackwell adds support for FP4 and FP6 data types with eighth precision floating point processing

Nvidia claims 20 petaflops of FP4 compute with the dual GB100 die B100 accelerator

Latest revision as of 04:07, 6 May 2024 (edit) Welcome (talk \| contribs) (Nvidia claims 20 petaflops of FP4 compute with the dual GB100 die B100 accelerator) Tags: Mobile web edit, Mobile edit ← Older edit	Latest revision as of 04:07, 6 May 2024 (edit) (undo) Welcome (talk \| contribs) (Nvidia claims 20 petaflops of FP4 compute with the dual GB100 die B100 accelerator) Tags: Mobile web edit, Mobile edit
(No difference)