sustcsonglin / flash-linear-attention
Efficient implementations of state-of-the-art linear attention models in Pytorch and Triton
☆1,190Updated this week
Related projects: ⓘ
- Implementation of 💍 Ring Attention, from Liu et al. at Berkeley AI, in Pytorch☆451Updated last month
- Transformers with Arbitrarily Large Context☆613Updated last month
- Puzzles for learning Triton☆966Updated this week
- A collection of AWESOME things about mixture-of-experts☆920Updated last month
- Helpful tools and examples for working with flex-attention☆341Updated last month
- Tile primitives for speedy kernels☆1,489Updated this week
- Annotated version of the Mamba paper☆445Updated 6 months ago
- Official PyTorch implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States☆975Updated 2 months ago
- Ring attention implementation with flash attention☆529Updated this week
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆452Updated last week
- A simple and efficient Mamba implementation in pure PyTorch and MLX.☆909Updated last week
- Repo for "Monarch Mixer: A Simple Sub-Quadratic GEMM-Based Architecture"☆530Updated 3 months ago
- Tutel MoE: An Optimized Mixture-of-Experts Implementation☆711Updated this week
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,099Updated 7 months ago
- Official Implementation of EAGLE-1 and EAGLE-2☆747Updated 3 weeks ago
- GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection☆1,354Updated last week
- Official JAX implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States☆343Updated last month
- depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.☆418Updated last week
- A simple and effective LLM pruning approach.☆614Updated last month
- A pytorch quantization backend for optimum☆758Updated this week
- [ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation☆568Updated 2 weeks ago
- Schedule-Free Optimization in PyTorch☆1,800Updated last month
- ☆1,164Updated last week
- FlashInfer: Kernel Library for LLM Serving☆1,138Updated this week
- A Pytorch implementation of Sparsely-Gated Mixture of Experts, for massively increasing the parameter count of language models☆610Updated last year
- Flash Attention in ~100 lines of CUDA (forward pass only)☆558Updated 5 months ago
- PyTorch native quantization and sparsity for training and inference☆726Updated this week
- Pipeline Parallelism for PyTorch☆708Updated 3 weeks ago
- Building blocks for foundation models.☆345Updated 8 months ago
- Minimalistic large language model 3D-parallelism training☆1,111Updated this week