On a B200, the nvjet_tst_16x64_64x16_4x1_v_bz_TNN kernel is used, and it takes roughly 8.1 microseconds. On a H200, the nvjet_tst_64x8_64x16_4x1_v_bz_TNT kernel is ...
Abstract: A mixed-precision analog compute-in-memory (Mix-ACIM) is presented for mixed-precision vector-matrix multiplication (VMM). The design features an all-analog current-domain fixed-point (FxP) ...
Abstract: The demand for high-speed matrix multiplication continues to grow due to recent developments in images processing, graphics processing, digital signal processing and communication via ...
There was an error while loading. Please reload this page.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback