GeezerVisonThe World Wide Web, restored to 800 pixels of glory. |
| Home|About|Gaming cabinet|Archives|How it works|Starter sites |
GeezerVison is showing github.com — Build software better, together
#
gpu-kernelHere are 13 public repositories matching this topic...Sort: Most stars
Sort options
TritonParse: A Compiler Tracer, Visualizer, and Reproducer for Triton Kernels
debugging compiler pytorch triton structured-logging interactive-visualization ir-analysis gpu-kernel ir-visualization
An end-to-end agent project for GPU kernel implementation, analysis, profiling, and iterative optimization. It helps an agent turn PyTorch logic or an existing kernel into a high-performance GPU kernel through a structured, profile-driven workflow.
agent skills gpu nvidia-gpu gpu-programming amd-gpu agentic-workflow self-improving-agent gpu-kernel autoresearch kernel-generation
可验证的 CUDA 学习主线：SGEMM、通用 GPU 算子、性能优化与轻量推理组件
learning education tutorial cplusplus hpc cuda stable gemm inference-engine gpu-programming sgemm cuda-programming ai-infra tensor-core kernel-optimization gpu-kernel
Wire the 1CatAI Split-D D256 FlashAttention kernel (fishlikeX/sm70-attn, MIT) into NInfer on Tesla V100 sm_70: +36-41% prefill, TTFT -3min, decode unchanged. Measured data + integration guide. Published by the user with AI assistance.
cuda volta prefill v100 inference-optimization kv-cache long-context llama-cpp llm-inference flash-attention d256 tesla-v100 qwen3 gpu-kernel sm70 attention-kernel 27b ninfer split-d splitkv
Inference Optimization by Economic Kernel Search
06 · 专精课 · 算子开发与优化（35 讲）：CUDA 与 Triton、Tensor Core、Blackwell FP4、GEMM 与 FlashAttention 实现
GQA/MQA-extended FlashAttention-2 in Triton, with a KV-cache decode kernel — built on a public tutorial baseline, hardened for correctness (4 bugs found/fixed), and benchmarked against PyTorch SDPA on NVIDIA L4.
Production-grade HIP kernel optimization lab — matrix ops, reductions, shared memory patterns
LayerNorm kernel implemented across three abstraction layers: CUDA C++, hand-written PTX (SM-120 Blackwell), and Ascend C (CANN 8.x). Demonstrates cross-architecture memory-bandwidth optimization.
cuda nvidia huawei cann systems-engineering memory-bandwidth ptx blackwell cross-architecture layernorm llm-inference low-level-optimization gpu-kernel sm-120 ascend-c
Explain Triton tensor shapes and coordinate provenance statically, without executing kernels. VS Code panel + MCP server for agents.
Hand-written Triton PagedAttention decode kernel with reproducible correctness & performance benchmarks against PyTorch baselines
寻找乐趣, 寻找生活 — ML / GPU Kernel / 技术笔记
Flash-attention forward kernel in Mojo for video DiT self-attention on RDNA3 (gfx1100)
Add this topic to your repoTo associate your repository with the gpu-kernel topic, visit your repo's landing page and select "manage topics." |
| You are visitor 01402503 — GeezerVison does not store the pages you look up. Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required. |