7nm

Parameterized N×N output-stationary systolic array accelerator for INT8 neural network inference. Full RTL-to-GDS flow on ASAP7 7nm using Cadence Genus + Innovus. 667 MHz, 42.7 GOPS peak throughput, 0.33 mW/GOP. SystemVerilog RTL, synthesis, place-and-route and self-checking testbench included.

𓁰 PtahCore — open-source FP8 tensor accelerator, RTL→GDSII on open 7nm ASAP7, 100% open toolchain

Transistor-level design and characterization in ASAP7 7nm FinFET: a dual-Vt domino AND8 (11% leakage reduction) and 6T vs 8T SRAM bitcells across read, write, and hold. Includes a measurement-setup discrepancy found in the original decks.

8 Phoenix: An open-source 7nm AI accelerator utilizing Silicon Photonics for massive-scale model inference. Features a deterministic dataflow architecture and 8-chip optical interconnect. 中文: 8 Phoenix (八凤)：一款基于硅光子技术的开源 7nm AI 加速器，专为大规模模型推理设计。采用确定性数据流架构及 8 芯片光互连方案。

This repository documents my work on 7nm FinFET circuit design and characterization using the ASAP7 PDK, Ngspice, and Xschem, including CMOS inverter and bandgap reference simulations