A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training
☆936Sep 3, 2026Updated this week
Alternatives and similar repositories for MagiAttention
Users that are interested in MagiAttention are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- A plug-and-play compiler that delivers free-lunch optimizations for both inference and training.☆332Updated this week
- Accelerating MoE with IO and Tile-aware Optimizations☆754Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,532Aug 12, 2026Updated 3 weeks ago
- MAGI-1: Autoregressive Video Generation at Scale☆3,779Jun 17, 2026Updated 2 months ago
- VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo☆2,187Updated this week
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,358Aug 28, 2025Updated last year
- ☆253Nov 19, 2025Updated 9 months ago
- [ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention☆709Jul 4, 2026Updated 2 months ago
- A unified inference and post-training framework for accelerated video generation.☆4,317Updated this week
- [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.☆1,043Feb 25, 2026Updated 6 months ago
- A Quirky Assortment of CuTe Kernels☆1,136Updated this week
- Ring attention implementation with flash attention☆1,050Sep 10, 2025Updated 11 months ago
- 🚀 Efficient implementations for emerging model architectures☆5,689Updated this week
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆690May 21, 2026Updated 3 months ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,519Updated this week
- xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism☆2,706Updated this week
- Helpful tools and examples for working with flex-attention☆1,234Updated this week
- FlashInfer: Kernel Library for LLM Serving☆6,326Updated this week
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,703Jan 17, 2026Updated 7 months ago
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆1,020Feb 5, 2026Updated 6 months ago
- ☆37Aug 7, 2025Updated last year
- Tile primitives for speedy kernels☆3,664Aug 28, 2026Updated last week
- flex-block-attn: an efficient block sparse attention computation library☆130Dec 26, 2025Updated 8 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- [NeurIPS 2025] Radial Attention: O(nlogn) Sparse Attention with Energy Decay for Long Video Generation☆607Nov 11, 2025Updated 9 months ago
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆135Jun 24, 2025Updated last year
- A sparse attention kernel supporting mix sparse patterns☆549Jan 18, 2026Updated 7 months ago
- DeepSeek-V3.2-Exp DSA Warmup Lightning Indexer training operator based on tilelang☆52Nov 19, 2025Updated 9 months ago
- Perplexity GPU Kernels☆605Nov 7, 2025Updated 9 months ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆545Updated this week
- high-performance linear attention kernel library built on TileLang☆685Aug 26, 2026Updated last week
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆200Feb 11, 2026Updated 6 months ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆7,341Updated this week
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- Allow torch tensor memory to be released and resumed later☆273Aug 21, 2026Updated last week
- A Survey of Efficient Attention Methods: Hardware-efficient, Sparse, Compact, and Linear Attention☆309Dec 1, 2025Updated 9 months ago
- Fast and Memory-Efficient Exact Attention (BF16/FP16/FP8/FP4) for Large Headdim, 1.5x~15x speedup over PyTorch SDPA.☆331Updated this week
- Efficient triton implementation of Native Sparse Attention.☆282May 23, 2025Updated last year
- Kernels, of the mega variety :)☆817May 26, 2026Updated 3 months ago
- SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention☆356Feb 24, 2026Updated 6 months ago
- A PyTorch-native inference engine with cache, parallelism, quantization and cpu offload for DiTs.☆1,269Updated this week