Skip to content
View DefTruth's full-sized avatar
๐ŸŽฏ
#pragma unroll
๐ŸŽฏ
#pragma unroll

Organizations

@vipshop @PaddlePaddle @xlite-dev

Block or report DefTruth

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
DefTruth/README.md

DefTruth | Trendshift

๐Ÿข Group: @xlite-dev | @vipshop | Prev. @PaddlePaddle ๐Ÿฐ

๐Ÿ›  Creator: xlite-dev | lite.ai.toolkit | Awesome-LLM-Inference | LeetCUDA | ffpa-attn ๐ŸŽง

๐Ÿ›  HGEMM | Awesome-DiT-Inference | lihang-notes(PDF, 200 Pages) | torchlm ๐ŸŽง

๐ŸŽ‰ Contributor: FastDeploy | vLLM | SGLang | Many Others โš™

๐Ÿค– Contact: [email protected] | GitHub: DefTruth | Zhihu(็ŸฅไนŽ): DefTruth ๐Ÿ“ž

โค I love open source, bro, and I think you do too. โค

Pinned Loading

  1. xlite-dev/LeetCUDA xlite-dev/LeetCUDA Public

    ๐Ÿ“šLeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners, 200+ CUDA & Tensor Cores Kernels, HGEMM, FA-2 MMA etc.๐Ÿ”ฅ

    Cuda 4.7k 508

  2. xlite-dev/lite.ai.toolkit xlite-dev/lite.ai.toolkit Public

    ๐Ÿ›  A lite C++ AI toolkit: 100+๐ŸŽ‰ models (Stable-Diffusion, FaceFusion, YOLO series, Det, Seg, Matting) with MNN, ORT and TensorRT.

    C++ 4.1k 743

  3. xlite-dev/Awesome-LLM-Inference xlite-dev/Awesome-LLM-Inference Public

    ๐Ÿ“šA curated list of Awesome LLM/VLM Inference Papers with codes: WINT8/4, FlashAttention, PagedAttention, Parallelism, MLA, etc.

    Python 4.1k 283

  4. PaddlePaddle/FastDeploy PaddlePaddle/FastDeploy Public

    Large Language Model Deployment Toolkit

    Cuda 3.2k 480

  5. xlite-dev/ffpa-attn xlite-dev/ffpa-attn Public

    ๐Ÿ“šFFPA(Split-D): Extend FlashAttention with Split-D for large headdim, O(1) GPU SRAM complexity, 1.8x~3xโ†‘๐ŸŽ‰ faster than SDPA EA.

    Cuda 185 8

  6. vllm-project/vllm vllm-project/vllm Public

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Python 49.3k 7.9k