DefTruth/CUDA-Learn-Notes
🎉CUDA 笔记 / 大模型手撕CUDA / C++笔记,更新随缘: flash_attn、sgemm、sgemv、warp reduce、block reduce、dot product、elementwise、softmax、layernorm、rmsnorm、hist etc.
CudaGPL-3.0
Stargazers
- Alwaysssssss
- BaofengZan
- BeverlyCrl
- blacksino
- ChunelFenghorizon, Ex Alibaba
- DefTruth@PaddlePaddle
- DIPTEShenZhen,China
- edsonke
- fdr27134
- GGgary666香港科技大学(广州)
- gm3g11University of Notre Dame
- gqjiaNEFU
- JailedBirdoppo
- Jameskrydtwave technology
- JiaoYanMoGuXiaomi Corporation
- jujimeizuoJiangnan University
- l-sf电子科技大学
- maliangzhibi
- MetaBluesBeijing, China
- mklf@bupt
- NALLEINBaidu
- OutBreak-hui
- piDackShanghai
- qdLMFCurrently unemployed, looking for a job in SLAM
- qpc001
- rlczddl
- ShenJunkunWuhan University
- SiriusDMHuazhong University of Science and Technology
- sofzh
- sonderlauHangzhou Dianzi University
- wjxzjuSJTU
- xinsuinizhuan
- yhwang-hub
- yisa2
- yuhengcai1
- ZonePGUSTC