openmlsys / openmlsys-cudaLinks
Tutorials for writing high-performance GPU operators in AI frameworks.
☆132Updated 2 years ago
Alternatives and similar repositories for openmlsys-cuda
Users that are interested in openmlsys-cuda are comparing it to the libraries listed below
Sorting:
- ☆141Updated last year
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆41Updated 7 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆223Updated 2 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆121Updated 5 months ago
- ☆139Updated last year
- A tutorial for CUDA&PyTorch☆158Updated 9 months ago
- Code base and slides for ECE408:Applied Parallel Programming On GPU.☆135Updated 4 years ago
- llm theoretical performance analysis tools and support params, flops, memory and latency analysis.☆108Updated 3 months ago
- ☆109Updated 6 months ago
- ☆150Updated 9 months ago
- 使用 CUDA C++ 实现的 llama 模型推理框架☆62Updated 11 months ago
- 分层解耦的深度学习推理引擎☆76Updated 8 months ago
- ☆174Updated 2 years ago
- 使用 cutlass 实现 flash-attention 精简版,具有教学意义☆50Updated last year
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆76Updated last year
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆45Updated 4 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated last month
- ☆210Updated 11 months ago
- ☆100Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency