Accelerating MoE with IO and Tile-aware Optimizations
☆630Apr 1, 2026Updated 2 weeks ago
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☆1,403Apr 10, 2026Updated last week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆5,497Updated this week
- ☆119May 19, 2025Updated 10 months ago
- FlashInfer: Kernel Library for LLM Serving☆5,372Apr 11, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,286Aug 28, 2025Updated 7 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- ☆38Aug 7, 2025Updated 8 months ago
- Tile primitives for speedy kernels☆3,312Apr 8, 2026Updated last week
- PyTorch bindings for CUTLASS grouped GEMM.☆186Apr 8, 2026Updated last week
- 🚀 Efficient implementations for emerging model architectures☆4,878Updated this week
- A Quirky Assortment of CuTe Kernels☆924Updated this week
- Perplexity GPU Kernels☆567Nov 7, 2025Updated 5 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…☆3,280Updated this week
- A lightweight design for computation-communication overlap.☆226Jan 20, 2026Updated 2 months ago
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆984Feb 5, 2026Updated 2 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆835Updated this week
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,057Sep 4, 2024Updated last year
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆98Dec 2, 2025Updated 4 months ago
- LM engine is a library for pretraining/finetuning LLMs☆164Updated this week
- ☆134Jun 6, 2025Updated 10 months ago
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆509Updated this week
- a size profiler for cuda binary☆70Jan 15, 2026Updated 3 months ago
- incubator repo for CUDA-TileIR backend☆127Mar 18, 2026Updated last month
- Efficient Long-context Language Model Training by Core Attention Disaggregation☆98Apr 7, 2026Updated last week
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- ☆22Aug 20, 2025Updated 7 months ago
- Nex Venus Communication Library☆73Nov 17, 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”☆166Apr 11, 2026Updated last week
- DeeperGEMM: crazy optimized version☆86May 5, 2025Updated 11 months ago
- A sparse attention kernel supporting mix sparse patterns☆497Jan 18, 2026Updated 3 months ago
- Distributed MoE in a Single Kernel [NeurIPS '25]☆247Apr 6, 2026Updated last week
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,296Jan 17, 2026Updated 3 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆380Jul 10, 2025Updated 9 months ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆536Feb 10, 2025Updated last year
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,200Updated this week
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆6,376Updated this week
- Analyze computation-communication overlap in V3/R1.☆1,149Mar 21, 2025Updated last year
- SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention☆295Feb 24, 2026Updated last month
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆177Feb 11, 2026Updated 2 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆826Mar 6, 2025Updated last year
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆260Feb 13, 2026Updated 2 months ago