Kedreamix / pytorch-cppcuda-tutorialLinks
tutorial for writing custom pytorch cpp+cuda kernel, applied on volume rendering (NeRF)
☆27Updated last year
Alternatives and similar repositories for pytorch-cppcuda-tutorial
Users that are interested in pytorch-cppcuda-tutorial are comparing it to the libraries listed below
Sorting:
- Implement Flash Attention using Cute.☆85Updated 5 months ago
- Tutorials for writing high-performance GPU operators in AI frameworks.☆130Updated last year
- ☆132Updated last month
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆36Updated 2 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆79Updated 3 weeks ago
- CPU Memory Compiler and Parallel programing☆26Updated 6 months ago
- ☆73Updated 3 weeks ago
- Implement custom operators in PyTorch with cuda/c++☆62Updated 2 years ago
- ☆85Updated 2 months ago
- ☆67Updated 7 months ago
- [ICLR 2025] Mixture Compressor for Mixture-of-Experts LLMs Gains More☆45Updated 3 months ago
- Codes & examples for "CUDA - From Correctness to Performance"☆98Updated 7 months ago
- Quantized Attention on GPU☆44Updated 6 months ago
- XAttention: Block Sparse Attention with Antidiagonal Scoring☆160Updated this week
- A tutorial for CUDA&PyTorch☆142Updated 4 months ago
- 📚FFPA(Split-D): Extend FlashAttention with Split-D for large headdim, O(1) GPU SRAM complexity, 1.8x~3x↑🎉 faster than SDPA EA.☆184Updated 3 weeks ago
- SGEMM optimization with cuda step by step☆19Updated last year
- Gensis is a lightweight deep learning framework written from scratch in Python, with Triton as its backend for high-performance computing…☆13Updated 5 months ago
- ☆48Updated this week
- Optimize softmax in triton in many cases☆21Updated 9 months ago
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆36Updated last month
- 使用 CUDA C++ 实现的 llama 模型推理框架☆57Updated 7 months ago
- Examples of CUDA implementations by Cutlass CuTe☆190Updated 4 months ago
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆51Updated 2 weeks ago
- llm theoretical performance analysis tools and support params, flops, memory and latency analysis.☆92Updated last week
- flash attention tutorial written in python, triton, cuda, cutlass☆370Updated 3 weeks ago
- Triton Documentation in Chinese Simplified / Triton 中文文档☆71Updated last month
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆100Updated last year
- ☆48Updated 3 weeks ago
- ☆134Updated last year