A plug-and-play compiler that delivers free-lunch optimizations for both inference and training.
☆337Nov 13, 2026Updated this week
Alternatives and similar repositories for MagiCompiler
Users that are interested in MagiCompiler 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 Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training☆974Updated this week
- FA4-based Relative Attention Kernel developed by TML and Colfax☆18Sep 11, 2026Updated 2 months ago
- ☆2,127Apr 11, 2026Updated 7 months ago
- MAGI-2-preview: Scaling Video Generation Models Efficiently☆658Aug 6, 2026Updated 3 months ago
- ☆97Feb 5, 2026Updated 9 months ago
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆209Feb 11, 2026Updated 9 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆697Sep 16, 2026Updated 2 months ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆562Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,580Sep 18, 2026Updated last month
- A Triton-only attention backend for vLLM☆28Jul 14, 2026Updated 4 months ago
- VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo☆2,294Updated this week
- NVIDIA cuTile learn☆168Dec 9, 2025Updated 11 months ago
- ☆11Oct 29, 2026Updated 2 weeks ago
- Tile-based language built for AI computation across all scales☆197Aug 25, 2026Updated 2 months ago
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- Code repository for the SOSP'25 paper DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism.☆23Nov 28, 2025Updated 11 months ago
- A Quirky Assortment of CuTe Kernels☆1,194Oct 30, 2026Updated 2 weeks ago
- flex-block-attn: an efficient block sparse attention computation library☆133Dec 26, 2025Updated 10 months ago
- mKernel: fast multi-node, multi-GPU fused kernels☆312Updated this week
- High-performance LLM operator library built on TileLang.☆203Updated this week
- MAGI-1: Autoregressive Video Generation at Scale☆3,806Jun 17, 2026Updated 5 months ago
- Unified tracing profiler and visualizer, one timeline from CPU to GPU/HES to in-kernel zones☆152Sep 21, 2026Updated last month
- Accelerating MoE with IO and Tile-aware Optimizations☆791Aug 29, 2026Updated 2 months ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,379Aug 28, 2025Updated last year
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,618Updated this week
- ☆54May 19, 2025Updated last year
- https://wavespeed.ai/ Context parallel attention that accelerates DiT model inference with dynamic caching☆436Jul 5, 2025Updated last year
- ☆147Updated this week
- Code and reproducibility artifacts for reconstructing megakernel persistence with CUDA PDL☆28Aug 5, 2026Updated 3 months ago
- A lightweight design for computation-communication overlap.☆253Jan 20, 2026Updated 9 months ago
- Official implementation of paper "VMoBA: Mixture-of-Block Attention for Video Diffusion Models"☆65Jul 1, 2025Updated last year
- [ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention☆712Jul 4, 2026Updated 4 months ago
- [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.☆1,268Feb 25, 2026Updated 8 months ago
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- FLA but cuTile☆27Apr 17, 2026Updated 7 months ago
- ☆37Aug 7, 2025Updated last year
- Piecewise-Taylor Attention☆48Aug 4, 2026Updated 3 months ago
- NCU-driven iterative optimization workflow for CUDA/CUTLASS/Triton/CuTe DSL kernels.☆25Apr 10, 2026Updated 7 months ago
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆52Jul 4, 2025Updated last year
- ☆32Jul 2, 2025Updated last year
- ☆44Sep 8, 2025Updated last year
Are these results useful?