Accelerating MoE with IO and Tile-aware Optimizations
☆664May 3, 2026Updated this week
Alternatives and similar repositories for sonic-moe
Users that are interested in sonic-moe are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Distributed Compiler based on Triton for Parallel Systems☆1,420Apr 22, 2026Updated 2 weeks ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆6,095Updated this week
- FlashInfer: Kernel Library for LLM Serving☆5,544May 2, 2026Updated last week
- A Quirky Assortment of CuTe Kernels☆955Updated this week
- ☆119May 19, 2025Updated 11 months ago
- 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.
- Tile primitives for speedy kernels☆3,336Apr 29, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,297Aug 28, 2025Updated 8 months ago
- ☆38Aug 7, 2025Updated 9 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆186Apr 8, 2026Updated last month
- 🚀 Efficient implementations for emerging model architectures☆5,032May 1, 2026Updated last week
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆995Feb 5, 2026Updated 3 months ago
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,312May 1, 2026Updated last week
- Perplexity GPU Kernels☆573Nov 7, 2025Updated 6 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆384Jul 10, 2025Updated 9 months ago
- 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.
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆855Updated this week
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,065Sep 4, 2024Updated last year
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆100Dec 2, 2025Updated 5 months ago
- A lightweight design for computation-communication overlap.☆229Jan 20, 2026Updated 3 months ago
- LM engine is a library for pretraining/finetuning LLMs☆166Updated this week
- ☆136Jun 6, 2025Updated 11 months ago
- From Minimal GEMM to Everything☆202Feb 10, 2026Updated 2 months ago
- A sparse attention kernel supporting mix sparse patterns☆509Jan 18, 2026Updated 3 months ago
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆528Updated this week
- 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.
- a size profiler for cuda binary☆70Jan 15, 2026Updated 3 months ago
- incubator repo for CUDA-TileIR backend☆134Apr 22, 2026Updated 2 weeks ago
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆98Apr 7, 2026Updated last month
- ☆23Aug 20, 2025Updated 8 months ago
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,241Updated this week
- Nex Venus Communication Library☆74Nov 17, 2025Updated 5 months ago
- ☆248Nov 19, 2025Updated 5 months ago
- Code for the papers: “Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling” and “Adaptive Block-Scaled Data Types”☆173Apr 21, 2026Updated 2 weeks ago
- DeeperGEMM: crazy optimized version☆86May 5, 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.
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆7,200Apr 24, 2026Updated 2 weeks ago
- Distributed MoE in a Single Kernel [NeurIPS '25]☆250Updated this week
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,342Jan 17, 2026Updated 3 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆151May 29, 2025Updated 11 months ago
- A throughput-oriented high-performance serving framework for LLMs☆956Mar 29, 2026Updated last month
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆543Feb 10, 2025Updated last year
- Analyze computation-communication overlap in V3/R1.☆1,152Mar 21, 2025Updated last year