Streamimg Table
===
- [2025/06/11](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/rySYWmVQgx)
- [2025/06/18](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/rJCwPaL7lg)
- [2025/07/02](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/ryKhVGpNex)
- [2025/07/09](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BJVo0BMrex)
- [2025/07/16](https://hackmd.io/tjvgXuZkTiqbSiEStE8awQ)
- [2025/07/23](https://hackmd.io/ZRd3rCr1QWeYEiw8I1Hb7g)
- [2025/07/30](https://hackmd.io/nPWCs2SySbuTJkAF4c1vqw)
- [2025/08/06](https://hackmd.io/DsCsHyjrTbq58s_id5mkNA?view)
- [2025/08/13](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Hy3Ia9Xdeg)
- [2025/08/20](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BkrqngzYel)
- [2025/09/04](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/B1OiY8kcxe)
- [2025/09/11](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Sk_DuHD9ll)
- [2025/09/18](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/SJ2okm4ieg)
- [2025/09/25](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HJ5tVrnoee)
- [2025/10/02](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/r1vuOJEhgl)
- [2025/10/09](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/H1kutsypgl)
- [2025/10/16](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HyRMM6O6lx)
- [2025/10/23](https://hackmd.io/1ZOTdtKyRdC0p08dVeyDNg)
- [2025/10/30](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Hk5OI7R0le)
- [2025/11/06](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/SylKhHARle)
- [2025/11/13](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/rkDw2ollWx)
- [2025/11/20](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/B1RtQsKeZg)
- [2025/11/27](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/r1DggtpxWx)
- [2025/12/05](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BJuUZppWWl)
- [2025/12/11](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/r1qxh68z-x)
- [2025/12/18](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BktEB90GWe)
- [2025/12/25](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/S1WGQFLQbe)
- [2026/01/08](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/ByI9Gc07Zg)
- [2026/01/15](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/S1_9SsiEZl)

筆記
---
- [RDMA](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HJBIpCcGgx)
- [G10](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/SJsENTAExg)
- [Forest](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/SJJOSmstgg)
- [FlashTensor](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BJmluD1Zee)
- [GPUVM & RDMA](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HJBIpCcGgx)
- [Mirage](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/B1X1Y470ee)
- [Tensor core slide(zhiyi)](https://docs.google.com/presentation/d/1ApnSvTWbfixNKhjzSrSNacuTrAsTMopUU7oXdmm8UFg/edit?usp=sharing)
- [FlexInfer: Breaking Memory Constraint via Flexible and Efficient Offloading for On-Device LLM Inference](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Hk9zddu-bg)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/ryRIoUKW-g)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Syrw6rhWZx)
- [SwapMoE: Serving Off-the-shelf MoE-based Large Language Models with Tunable Memory Budget](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/rkxACTQmfZg)
- [fineMoE: Taming Latency-Memory Trade-Off in MoE-Based LLM Serving via Fine-Grained Expert Offloading](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/SkkO1Jwf-x)
- [MOE-INFINITY: Efficient MoE Inference on Personal Machines with Sparsity-Aware Expert Cache](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/H1260kymbe)
- [FIDDLER: CPU-GPU ORCHESTRATION FOR FAST INFERENCE OF MIXTURE-OF-EXPERTS MODELS](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Hk31b76zWx)
- [KTransformers: Unleashing the Full Potential of CPU/GPU Hybrid Inference for MoE Models](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/ryDOrgnGZe)
- [ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/BykwtMd7-e)
-

資源
---
- [UVMsmart build](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/Hk18JMuqgx)
- [Cublas run note](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HyYnuLGBee)
- [Cutlass Nsight system / Nsight compute note](https://hackmd.io/d3N9GbDuS8isQYs7WXxrog)
- [Pytorch model to Onnx-sim](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/HJfTeuX0xe)
- [Cuda library / Python Binding安裝方法](https://hackmd.io/@jQmycuO_SgeGjVC0cL-RTw/rySfjYjAlg)