- Hsinchu, Taiwan
Highlights
- Pro
Pinned Loading
-
-
2025-Open-hackthon-NDVIgnition
2025-Open-hackthon-NDVIgnition PublicGPU-accelerated aerial photogrammetry pipeline — NVIDIA/NCHC Open Hackathon 2025, 2nd place (5.36× end-to-end speedup)
Python
-
gpu-byteplane-perf
gpu-byteplane-perf PublicByte-plane scan and progressive aggregation CUDA kernels on H200: memory-access optimization with Nsight Compute evidence
Cuda
-
-
moe-serving-lab
moe-serving-lab PublicLarge-MoE LLM serving measurements on Hopper and Blackwell (FP8/NVFP4), plus preregistered inference case studies
Python
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.

