Pinned Loading
-
video-generation-model
video-generation-model PublicImplementing and training a video generation model, based on DIT
Python 1
-
cnn_chip
cnn_chip PublicA tiny Verilog CNN accelerator for the Tang Nano 20K. Features INT8 quantized weights, a streaming MAC array, and full on-chip MNIST inference.
Verilog
-
-
-
gpu_design
gpu_design PublicA from-scratch GPU stack: custom SIMT hardware in Verilog → a 16-bit ISA + assembler → J++, a small C-like language and compiler — all running on a Tang Nano 20K FPGA.
SystemVerilog
-
nano-inference
nano-inference PublicCUDA LLM engine that beats llama.cpp 1.8× on a Jetson Orin ▎ Nano
Cuda
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.



