Alternatives and similar repositories for MagiAttention

Users that are interested in MagiAttention are comparing it to the libraries listed below

• thu-ml / SpargeAttn
  [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
  ☆746Updated last week
• mit-han-lab / Block-Sparse-Attention
  A sparse attention kernel supporting mix sparse patterns
  ☆322Updated 8 months ago
• feifeibear / long-context-attention
  USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference
  ☆582Updated last week
• svg-project / Sparse-VideoGen
  [ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention
  ☆517Updated 3 weeks ago
• vipshop / cache-dit
  A Unified Cache Acceleration Framework for 🤗Diffusers: Qwen-Image-Lightning, Qwen-Image, HunyuanImage, Wan, FLUX, etc.
  ☆421Updated this week
• stepfun-ai / Step3
  ☆427Updated 2 months ago
• thu-nics / DiTFastAttn
  ☆182Updated 9 months ago
• mit-han-lab / distrifuser
  [CVPR 2024 Highlight] DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models
  ☆708Updated 10 months ago
• xlite-dev / Awesome-DiT-Inference
  📚A curated list of Awesome Diffusion Inference Papers with Codes: Sampling, Cache, Quantization, Parallelism, etc.🎉
  ☆429Updated 2 months ago
• XunhaoLai / native-sparse-attention-triton
  Efficient triton implementation of Native Sparse Attention.
  ☆238Updated 5 months ago
• thu-ml / SLA
  SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention
  ☆113Updated this week
• xdit-project / DistVAE
  A parallelism VAE avoids OOM for high resolution image generation
  ☆81Updated 2 months ago
• NVlabs / Fast-dLLM
  Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"
  ☆604Updated this week
• ByteDance-Seed / VeOmni
  VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo
  ☆1,231Updated this week
• mit-han-lab / radial-attention
  [NeurIPS 2025] Radial Attention: O(nlogn) Sparse Attention with Energy Decay for Long Video Generation
  ☆531Updated last month
• mit-han-lab / x-attention
  [ICML 2025] XAttention: Block Sparse Attention with Antidiagonal Scoring
  ☆239Updated 3 months ago
• zhuzilin / ring-flash-attention
  Ring attention implementation with flash attention
  ☆903Updated last month
• nunchaku-tech / deepcompressor
  Model Compression Toolbox for Large Language Models and Diffusion Models
  ☆678Updated 2 months ago
• sandyresearch / chipmunk
  🎬 3.7× faster video generation E2E 🖼️ 1.6× faster image generation E2E ⚡ ColumnSparseAttn 9.3× vs FlashAttn‑3 💨 ColumnSparseGEMM 2.5× …
  ☆87Updated last month
• fla-org / native-sparse-attention
  🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"
  ☆903Updated 7 months ago
• zhijie-group / Discrete-Diffusion-Forcing
  Discrete Diffusion Forcing (D2F): dLLMs Can Do Faster-Than-AR Inference
  ☆171Updated last month
• fla-org / flame
  🔥 A minimal training framework for scaling FLA models
  ☆266Updated last month
• NVlabs / COAT
  [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training
  ☆242Updated 2 months ago
• chengzeyi / ParaAttention
  https://wavespeed.ai/ Context parallel attention that accelerates DiT model inference with dynamic caching
  ☆385Updated 3 months ago
• mit-han-lab / duo-attention
  [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
  ☆496Updated 8 months ago
• alexzhang13 / flashattention2-custom-mask
  Triton implementation of FlashAttention2 that adds Custom Masks.
  ☆141Updated last year
• Tele-AI / TeleTron
  To pioneer training long-context multi-modal transformer models
  ☆59Updated 2 months ago
• Shenyi-Z / TaylorSeer
  [ICCV2025] From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers
  ☆307Updated 2 months ago
• ByteDance-Seed / ByteCheckpoint
  ByteCheckpoint: An Unified Checkpointing Library for LFMs
  ☆249Updated 3 months ago
• mit-han-lab / patch_conv
  Patch convolution to avoid large GPU memory usage of Conv2D
  ☆92Updated 9 months ago