MSLK (Meta Superintelligence Labs Kernels) is a collection of PyTorch GPU operator libraries that are designed and optimized for GenAI training and inference, such as FP8 row-wise quantization and collective communications.
☆160Oct 30, 2026Updated this week
Alternatives and similar repositories for MSLK
Users that are interested in MSLK are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Autonomous GPU kernel optimization system driven by AI agents.☆32Mar 29, 2026Updated 7 months ago
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆977Updated this week
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆375Updated this week
- Triton-based Symmetric Memory operators and examples☆113May 15, 2026Updated 5 months ago
- Easy recipes to speed up latency of Flux, QwenImage, and LTX-2 with NVFP4 and MXFP8 on Blackwell.☆14Apr 10, 2026Updated 6 months ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- Github mirror of trition-lang/triton repo.☆205Updated this week
- (deprecated) torchcomms: a modern PyTorch communications API☆403Updated this week
- FA4-based Relative Attention Kernel developed by TML and Colfax☆18Sep 11, 2026Updated last month
- ☆171Dec 27, 2024Updated last year
- High performance RMSNorm Implement by using SM Core Storage(Registers and Shared Memory)☆31Jan 22, 2026Updated 9 months ago
- A Quirky Assortment of CuTe Kernels☆1,189Updated this week
- torch_remat fine-grained activation checkpointing API☆23Updated this week
- A dynamic binary instrumentation tool for tracing and analyzing GPU kernel instructions.☆84Oct 27, 2026Updated last week
- NVFP4 Flash-Attention 4 on BlackWell☆69Sep 26, 2026Updated last month
- 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.
- An Optimizer for Nvidia Compilers.☆146Oct 21, 2026Updated 2 weeks ago
- Unified tracing profiler and visualizer, one timeline from CPU to GPU/HES to in-kernel zones☆150Sep 21, 2026Updated last month
- Ship correct and fast LLM kernels to PyTorch☆156Jan 14, 2026Updated 9 months ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆561Updated this week
- Fast low-bit matmul kernels in Triton☆491Oct 1, 2026Updated last month
- Ulysses sequence-parallel all-to-all as a torch custom op, moved by the GPU copy engines into torch symmetric memory. Zero SM usage; 1.66…☆54Aug 25, 2026Updated 2 months ago
- a size profiler for cuda binary☆74Oct 3, 2026Updated last month
- A Triton JIT runtime and ffi provider in C++☆41Sep 24, 2026Updated last month
- An experimental communicating attention kernel based on DeepEP.☆34Jul 29, 2025Updated last year
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- Helpful tools and examples for working with flex-attention☆1,263Updated this week
- ☆37Aug 7, 2025Updated last year
- TritonParse: A Compiler Tracer, Visualizer, and Reproducer for Triton Kernels☆229Oct 20, 2026Updated 2 weeks ago
- Benchmark tests supporting the TiledCUDA library.☆19Nov 19, 2024Updated last year
- A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")☆407May 31, 2026Updated 5 months ago
- Distributed Compiler and Optimized Parallel Kernels☆1,576Sep 18, 2026Updated last month
- FLA but cuTile☆27Apr 17, 2026Updated 6 months ago
- QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning☆204Sep 2, 2026Updated 2 months ago
- Applied AI experiments and examples for PyTorch☆322Aug 22, 2025Updated last year
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Kernel Library for Large Headdim Attention (64~1024, BF16/FP8/FP4), 1.5x~15x↑ vs PyTorch SDPA.☆345Oct 23, 2026Updated last week
- Kernels, of the mega variety :)☆875May 26, 2026Updated 5 months ago
- An experimental implementation of compiler-driven automatic sharding of models across a given device mesh.☆98Oct 21, 2026Updated last week
- Accelerating MoE with IO and Tile-aware Optimizations☆787Aug 29, 2026Updated 2 months ago
- NCU-driven iterative optimization workflow for CUDA/CUTLASS/Triton/CuTe DSL kernels.☆25Apr 10, 2026Updated 6 months ago
- This module defines a type system for distributed training code, based off of JAX's sharding in types, but adapted for the PyTorch ecosys…☆46Oct 23, 2026Updated last week
- Claude Code agent skills for writing optimized Triton GPU kernels☆15Mar 6, 2026Updated 7 months ago