MagellaX / StreamAttnLinks
A high-performance attention mechanism that computes softmax normalization in a single streaming pass using running accumulators (online softmax).
☆26Updated last month
Alternatives and similar repositories for StreamAttn
Users that are interested in StreamAttn are comparing it to the libraries listed below
Sorting:
- Learning about CUDA by writing PTX code.☆138Updated last year
- pytorch from scratch in pure C/CUDA and python☆41Updated last year
- Tensor library with autograd using only Rust's standard library☆69Updated last year
- Quantized LLM training in pure CUDA/C++.☆180Updated this week
- SIMD quantization kernels☆87Updated last month
- PTX-Tutorial Written Purely By AIs (Deep Research of Openai and Claude 3.7)☆66Updated 6 months ago
- Extensive introductory writeup on Zig language functionalities☆10Updated last year
- Rust Implementation of micrograd☆53Updated last year
- PCCL (Prime Collective Communications Library) implements fault tolerant collective communications over IP☆126Updated 3 weeks ago
- Simple Transformer in Jax☆139Updated last year
- A really tiny autograd engine☆95Updated 4 months ago
- Here's all my Python/Numba (CUDA) code for the encoder block I made :)☆68Updated 5 months ago
- A simple MLX implementation for pretraining LLMs on Apple Silicon.☆83Updated last month
- could we make an ml stack in 100,000 lines of code?☆46Updated last year
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆105Updated 7 months ago
- ☆78Updated 2 weeks ago
- In this repository I have a code and brief explanations of the attempts that I made at the ARC-AGI (2024) challenges :)☆24Updated 10 months ago
- moondream in zig.☆74Updated 4 months ago
- look how they massacred my boy☆63Updated 11 months ago
- Experimental GPU language with meta-programming☆23Updated last year
- NanoGPT-speedrunning for the poor T4 enjoyers☆72Updated 5 months ago
- An implement of deep learning framework and models in C☆48Updated 6 months ago
- ☆96Updated last year
- My submission for the GPUMODE/AMD fp8 mm challenge☆29Updated 4 months ago
- Because it's there.☆16Updated last year
- Learnings and programs related to CUDA☆419Updated 3 months ago
- in this repository, i'm going to implement increasingly complex llm inference optimizations☆68Updated 4 months ago
- Dion optimizer algorithm☆361Updated last week
- Gradient descent is cool and all, but what if we could delete it?☆104Updated last month
- A graph visualization of attention☆57Updated 4 months ago