Implementation of FlashAttention (FA1-FA4) in PyTorch for educational and algorithmic clarity
☆220Apr 12, 2026Updated 2 months ago
Alternatives and similar repositories for FlashAttention-PyTorch
Users that are interested in FlashAttention-PyTorch are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Get down and dirty with FlashAttention2.0 in pytorch, plug in and play no complex CUDA kernels☆115Jul 31, 2023Updated 2 years ago
- flash attention tutorial written in python, triton, cuda, cutlass☆522Jan 20, 2026Updated 4 months ago
- Flash Attention in ~100 lines of CUDA (forward pass only)☆1,158Dec 30, 2024Updated last year
- a minimal cache manager for PagedAttention, on top of llama3.☆146Aug 26, 2024Updated last year
- Implement Flash Attention using Cute.☆108Dec 17, 2024Updated last year
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- Terminal UI for NVIDIA Nsight Systems profiles — timeline viewer, kernel navigator, NVTX hierarchy☆58Jun 12, 2026Updated last week
- ☆16Mar 13, 2023Updated 3 years ago
- Prune transformer layers☆74May 30, 2024Updated 2 years ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆92Jul 17, 2025Updated 11 months ago
- Triton implementation of FlashAttention2 that adds Custom Masks.☆176Aug 14, 2024Updated last year
- ☆27Aug 5, 2022Updated 3 years ago
- Implementation of Speculative Sampling as described in "Accelerating Large Language Model Decoding with Speculative Sampling" by Deepmind☆111Feb 29, 2024Updated 2 years ago
- triton ver of gqa flash attn, based on the tutorial☆12Aug 4, 2024Updated last year
- Fast inference from large lauguage models via speculative decoding☆916Aug 22, 2024Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Code for Blog Post: Can Better Cold-Start Strategies Improve RL Training for LLMs?☆20Mar 9, 2025Updated last year
- [ICLR 2025 & COLM 2025] Official PyTorch implementation of the Forgetting Transformer and Adaptive Computation Pruning☆149Feb 25, 2026Updated 3 months ago
- ☆16Nov 14, 2022Updated 3 years ago
- [ACL 2025] Squeezed Attention: Accelerating Long Prompt LLM Inference☆59Nov 20, 2024Updated last year
- ☆52May 19, 2025Updated last year
- Fast and memory-efficient exact attention☆24,170Updated this week
- simplest online-softmax notebook for explain Flash Attention☆17Jan 27, 2026Updated 4 months ago
- ☆35Dec 22, 2025Updated 5 months ago
- ☆46May 24, 2025Updated last year
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- [NeurIPS 2023 spotlight] Official implementation of HGRN in our NeurIPS 2023 paper - Hierarchically Gated Recurrent Neural Network for Se…☆68Apr 24, 2024Updated 2 years ago
- 🚀 Efficient implementations for emerging model architectures☆5,227Jun 11, 2026Updated last week
- High Performance FP8 GEMM Kernels for SM89 and later GPUs.☆21Jan 24, 2025Updated last year
- Implementation for POET and POET-X for LLM pretraining☆36Jun 9, 2026Updated last week
- Official Implementation for the ICLR2023 paper "Fuzzy Alignments in Directed Acyclic Graph for Non-autoregressive Machine Translation"☆14Mar 1, 2023Updated 3 years ago
- The evaluation framework for training-free sparse attention in LLMs☆123Jan 27, 2026Updated 4 months ago
- Standalone Flash Attention v2 kernel without libtorch dependency☆113Sep 10, 2024Updated last year
- Puzzles for learning Triton☆2,491Apr 1, 2026Updated 2 months ago
- Codes and data for ACL 2023 Findings paper "Click: Controllable Text Generation with Sequence Likelihood Contrastive Learning"☆18Feb 26, 2024Updated 2 years ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Code for the paper: https://arxiv.org/pdf/2309.06979.pdf☆21Jul 29, 2024Updated last year
- Recreating PyTorch from scratch (C/C++, CUDA, NCCL and Python, with multi-GPU support and automatic differentiation!)☆166Nov 25, 2025Updated 6 months ago
- Using FlexAttention to compute attention with different masking patterns☆47Sep 22, 2024Updated last year
- Llama causal LM fully recreated in LibTorch. Designed to be used in Unreal Engine 5☆16Sep 19, 2024Updated last year
- Flash Attention in ~100 lines of CUDA (forward pass only)☆12Jun 10, 2024Updated 2 years ago
- ☆40Dec 14, 2025Updated 6 months ago
- Use the tokenizer in parallel to achieve superior acceleration☆20Mar 21, 2024Updated 2 years ago