https://supercharleszhu.github.io/posts/2025-08-07-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2025-08-14-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-04-10-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-04-24-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-04-27-00/ 2026-08-24T10:38:00+00:00 https://supercharleszhu.github.io/posts/2026-04-28-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-04-30-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-08-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-10-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-11-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-12-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-14-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-05-19-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-04-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-05-01/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-08-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-08-01/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-26-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-06-29-00/ 2026-08-11T06:26:37+00:00 https://supercharleszhu.github.io/posts/2026-08-11-00/ 2026-08-11T07:00:00+00:00 https://supercharleszhu.github.io/categories/ 2026-08-24T10:38:41+00:00 https://supercharleszhu.github.io/tags/ 2026-08-24T10:38:41+00:00 https://supercharleszhu.github.io/archives/ 2026-08-24T10:38:41+00:00 https://supercharleszhu.github.io/about/ 2026-08-24T10:38:41+00:00 https://supercharleszhu.github.io/ https://supercharleszhu.github.io/tags/pytorch/ https://supercharleszhu.github.io/tags/torch-compile/ https://supercharleszhu.github.io/tags/triton/ https://supercharleszhu.github.io/tags/optimization/ https://supercharleszhu.github.io/tags/deep-learning/ https://supercharleszhu.github.io/tags/cuda/ https://supercharleszhu.github.io/tags/cutlass/ https://supercharleszhu.github.io/tags/kernel/ https://supercharleszhu.github.io/tags/nccl/ https://supercharleszhu.github.io/tags/distributed-training/ https://supercharleszhu.github.io/tags/torchrun/ https://supercharleszhu.github.io/tags/gpu/ https://supercharleszhu.github.io/tags/cute/ https://supercharleszhu.github.io/tags/gemm/ https://supercharleszhu.github.io/tags/hopper/ https://supercharleszhu.github.io/tags/blackwell/ https://supercharleszhu.github.io/tags/tma/ https://supercharleszhu.github.io/tags/wgmma/ https://supercharleszhu.github.io/tags/tcgen05/ https://supercharleszhu.github.io/tags/vllm/ https://supercharleszhu.github.io/tags/llm/ https://supercharleszhu.github.io/tags/inference/ https://supercharleszhu.github.io/tags/multiprocessing/ https://supercharleszhu.github.io/tags/zmq/ https://supercharleszhu.github.io/tags/kv-cache/ https://supercharleszhu.github.io/tags/paged-attention/ https://supercharleszhu.github.io/tags/sglang/ https://supercharleszhu.github.io/tags/radix-attention/ https://supercharleszhu.github.io/tags/inductor/ https://supercharleszhu.github.io/tags/fx/ https://supercharleszhu.github.io/tags/codegen/ https://supercharleszhu.github.io/tags/torch-compile/ https://supercharleszhu.github.io/tags/dependencies/ https://supercharleszhu.github.io/tags/scheduler/ https://supercharleszhu.github.io/tags/buffer-lifetime/ https://supercharleszhu.github.io/tags/qwen3/ https://supercharleszhu.github.io/tags/transformers/ https://supercharleszhu.github.io/tags/attention/ https://supercharleszhu.github.io/tags/gqa/ https://supercharleszhu.github.io/tags/rope/ https://supercharleszhu.github.io/tags/rmsnorm/ https://supercharleszhu.github.io/tags/swiglu/ https://supercharleszhu.github.io/tags/llm-architecture/ https://supercharleszhu.github.io/tags/dynamo/ https://supercharleszhu.github.io/tags/dynamic-shapes/ https://supercharleszhu.github.io/tags/piecewise-compile/ https://supercharleszhu.github.io/tags/llm-serving/ https://supercharleszhu.github.io/tags/ddp/ https://supercharleszhu.github.io/tags/distributeddataparallel/ https://supercharleszhu.github.io/tags/reducer/ https://supercharleszhu.github.io/tags/allreduce/ https://supercharleszhu.github.io/tags/broadcast-coalesced/ https://supercharleszhu.github.io/tags/ddp-optimizer/ https://supercharleszhu.github.io/tags/compiled-autograd/ https://supercharleszhu.github.io/tags/huggingface/ https://supercharleszhu.github.io/tags/trainer/ https://supercharleszhu.github.io/tags/dataloader/ https://supercharleszhu.github.io/tags/accelerate/ https://supercharleszhu.github.io/tags/lengthgroupedsampler/ https://supercharleszhu.github.io/tags/num-items-in-batch/ https://supercharleszhu.github.io/tags/cross-entropy/ https://supercharleszhu.github.io/tags/lm-head/ https://supercharleszhu.github.io/tags/fused-kernel/ https://supercharleszhu.github.io/tags/liger/ https://supercharleszhu.github.io/tags/verl/ https://supercharleszhu.github.io/tags/online-softmax/ https://supercharleszhu.github.io/tags/gpu-memory/ https://supercharleszhu.github.io/tags/rl/ https://supercharleszhu.github.io/tags/reinforcement-learning/ https://supercharleszhu.github.io/tags/grpo/ https://supercharleszhu.github.io/tags/ppo/ https://supercharleszhu.github.io/tags/deepseek-math/ https://supercharleszhu.github.io/tags/llm-training/ https://supercharleszhu.github.io/tags/fsdp/ https://supercharleszhu.github.io/tags/ray/ https://supercharleszhu.github.io/tags/kl-divergence/ https://supercharleszhu.github.io/tags/k3-estimator/ https://supercharleszhu.github.io/tags/memory-profiling/ https://supercharleszhu.github.io/tags/llm-inference/ https://supercharleszhu.github.io/tags/gpu-memory-utilization/ https://supercharleszhu.github.io/tags/model-serving/ https://supercharleszhu.github.io/tags/moe/ https://supercharleszhu.github.io/tags/mixture-of-experts/ https://supercharleszhu.github.io/tags/expert-parallelism/ https://supercharleszhu.github.io/tags/megatron/ https://supercharleszhu.github.io/tags/deepseek/ https://supercharleszhu.github.io/tags/deepep/ https://supercharleszhu.github.io/tags/all-to-all/ https://supercharleszhu.github.io/tags/llm-systems/ https://supercharleszhu.github.io/tags/nvshmem/ https://supercharleszhu.github.io/tags/pgas/ https://supercharleszhu.github.io/tags/symmetric-memory/ https://supercharleszhu.github.io/tags/ibgda/ https://supercharleszhu.github.io/tags/ibrc/ https://supercharleszhu.github.io/tags/tensor-parallelism/ https://supercharleszhu.github.io/tags/data-parallelism/ https://supercharleszhu.github.io/tags/parallelism/ https://supercharleszhu.github.io/tags/tvm-ffi/ https://supercharleszhu.github.io/tags/dlpack/ https://supercharleszhu.github.io/tags/abi/ https://supercharleszhu.github.io/tags/native-extension/ https://supercharleszhu.github.io/tags/kernel-library/ https://supercharleszhu.github.io/tags/python-wheels/ https://supercharleszhu.github.io/tags/cpp/ https://supercharleszhu.github.io/tags/miles/ https://supercharleszhu.github.io/tags/rollout/ https://supercharleszhu.github.io/tags/reward-model/ https://supercharleszhu.github.io/tags/megatron-lm/ https://supercharleszhu.github.io/tags/megatron-core/ https://supercharleszhu.github.io/tags/megatron-bridge/ https://supercharleszhu.github.io/tags/qwen/ https://supercharleszhu.github.io/tags/fine-tuning/ https://supercharleszhu.github.io/tags/sft/ https://supercharleszhu.github.io/tags/pipeline-parallelism/ https://supercharleszhu.github.io/tags/dataset/ https://supercharleszhu.github.io/categories/machine-learning-performance/ https://supercharleszhu.github.io/categories/torch-compile/ https://supercharleszhu.github.io/categories/kernel/ https://supercharleszhu.github.io/categories/communication/ https://supercharleszhu.github.io/categories/llm-inference/ https://supercharleszhu.github.io/categories/llm-architecture/ https://supercharleszhu.github.io/categories/distributed-training/ https://supercharleszhu.github.io/categories/machine-learning-training-framework/ https://supercharleszhu.github.io/categories/reinforcement-learning/ https://supercharleszhu.github.io/categories/machine-learning-llm-architecture/ https://supercharleszhu.github.io/categories/machine-learning-llm-inference/ https://supercharleszhu.github.io/categories/compiler/ https://supercharleszhu.github.io/page2/