MagellaX / StreamAttnLinks
A high-performance attention mechanism that computes softmax normalization in a single streaming pass using running accumulators (online softmax).
☆28Updated 3 months ago
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.☆151Updated last year
- SIMD quantization kernels☆92Updated 4 months ago
- PTX-Tutorial Written Purely By AIs (Deep Research of Openai and Claude 3.7)☆66Updated 9 months ago
- PCCL (Prime Collective Communications Library) implements fault tolerant collective communications over IP☆141Updated 3 months ago
- ☆84Updated 3 weeks ago
- Quantized LLM training in pure CUDA/C++.☆230Updated this week
- pytorch from scratch in pure C/CUDA and python☆41Updated last year
- Tensor library with autograd using only Rust's standard library☆71Updated last year
- in this repository, i'm going to implement increasingly complex llm inference optimizations☆79Updated 7 months ago
- coding CUDA everyday!☆72Updated last month
- ☆88Updated 2 months ago
- My submission for the GPUMODE/AMD fp8 mm challenge☆29Updated 7 months ago
- CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning☆294Updated this week
- Simple Transformer in Jax☆140Updated last year
- Write a fast kernel and run it on Discord. See how you compare against the best!☆66Updated 3 weeks ago
- NanoGPT (124M) quality in 2.67B tokens☆28Updated 3 months ago
- tiny code to access tenstorrent blackhole☆61Updated 7 months ago
- Official CLI and Python SDK for Prime Intellect - access GPU compute, remote sandboxes, RL environments, and distributed training infrast…☆133Updated this week
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆109Updated 10 months ago
- Official Problem Sets / Reference Kernels for the GPU MODE Leaderboard!☆182Updated 2 weeks ago
- Custom PTX Instruction Benchmark☆137Updated 10 months ago
- Rust Implementation of micrograd☆53Updated last year
- Hand-Rolled GPU communications library☆76Updated last month
- NanoGPT-speedrunning for the poor T4 enjoyers☆73Updated 8 months ago
- NSA Triton Kernels written with GPT5 and Opus 4.1☆70Updated 4 months ago
- A really tiny autograd engine☆98Updated 7 months ago
- MoE training for Me and You and maybe other people☆315Updated last week
- A simple MLX implementation for pretraining LLMs on Apple Silicon.☆85Updated 4 months ago
- Simple high-throughput inference library☆155Updated 7 months ago
- High-Performance SGEMM on CUDA devices☆115Updated 11 months ago