IHP 130nm ASIC tapeout of a 2x2 bfloat16 matrix matrix multiplication with DFT infrastructure. Iteration on the previous accelerator taped out on GF180.
-
Updated
Jul 10, 2026 - Verilog
IHP 130nm ASIC tapeout of a 2x2 bfloat16 matrix matrix multiplication with DFT infrastructure. Iteration on the previous accelerator taped out on GF180.
RTL implementation of a performance/area optimized bfloat16 adder and multiplication.
Verilog implementation of a Newton-Raphson floating-point compute unit on the Intel MAX10 FPGA supporting reciprocal, square root, division, and natural logarithm in bfloat16 format.
To associate your repository with the bfloat16 topic, visit your repo's landing page and select "manage topics."