🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"
☆1,006Feb 5, 2026Updated 4 months ago
Alternatives and similar repositories for native-sparse-attention
Users that are interested in native-sparse-attention are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Implementation of the sparse attention pattern proposed by the Deepseek team in their "Native Sparse Attention" paper☆808Aug 15, 2025Updated 10 months ago
- Efficient triton implementation of Native Sparse Attention.☆281May 23, 2025Updated last year
- 🚀 Efficient implementations for emerging model architectures☆5,279Updated this week
- MoBA: Mixture of Block Attention for Long-Context LLMs☆2,129Apr 3, 2025Updated last year
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆133Jun 24, 2025Updated last year
- 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 sparse attention kernel supporting mix sparse patterns☆528Jan 18, 2026Updated 5 months ago
- qwen-nsa☆87Oct 14, 2025Updated 8 months ago
- Muon is Scalable for LLM Training☆1,496Aug 3, 2025Updated 10 months ago
- Distributed Compiler based on Triton for Parallel Systems☆1,466Jun 25, 2026Updated last week
- Quantized Attention on GPU☆44Nov 22, 2024Updated last year
- Helpful tools and examples for working with flex-attention☆1,205Updated this week
- FlashInfer: Kernel Library for LLM Serving☆5,867Updated this week
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆538Feb 10, 2025Updated last year
- A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training☆855Jun 25, 2026Updated last week
- Deploy open-source AI quickly and easily - Special Bonus Offer • AdRunpod Hub is built for open source. One-click deployment and autoscaling endpoints without provisioning your own infrastructure.
- [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.☆1,002Feb 25, 2026Updated 4 months ago
- 🔥 A minimal training framework for scaling FLA models☆396Apr 22, 2026Updated 2 months ago
- Ring attention implementation with flash attention☆1,025Sep 10, 2025Updated 9 months ago
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,220Apr 8, 2026Updated 2 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆397Jul 10, 2025Updated 11 months ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆6,552Updated this week
- ☆117Feb 25, 2025Updated last year
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆845Mar 6, 2025Updated last year
- DeeperGEMM: crazy optimized version☆86May 5, 2025Updated last year
- 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.
- DeepSeek Native Sparse Attention pytorch implementation☆119Dec 17, 2025Updated 6 months ago
- ☆65Apr 26, 2025Updated last year
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆264Aug 9, 2025Updated 10 months ago
- Efficient Triton Kernels for LLM Training☆6,456Jun 23, 2026Updated last week
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆169Oct 13, 2025Updated 8 months ago
- ☆250Nov 19, 2025Updated 7 months ago
- ☆52May 19, 2025Updated last year
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,332Aug 28, 2025Updated 10 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆675May 21, 2026Updated last month
- 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.
- Tile primitives for speedy kernels☆3,497Jun 15, 2026Updated 2 weeks ago
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆770Aug 6, 2025Updated 10 months ago
- Efficient LLM Inference over Long Sequences☆394Jun 25, 2025Updated last year
- flash attention tutorial written in python, triton, cuda, cutlass☆525Jan 20, 2026Updated 5 months ago
- Fast low-bit matmul kernels in Triton☆473May 15, 2026Updated last month
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆7,422Jun 24, 2026Updated last week
- A bidirectional pipeline parallelism algorithm for computation-communication overlap in DeepSeek V3/R1 training.☆2,972Jan 14, 2026Updated 5 months ago