An efficient implementation of the NSA (Native Sparse Attention) kernel
☆133Jun 24, 2025Updated 10 months ago
Alternatives and similar repositories for nsa-release
Users that are interested in nsa-release are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆15Mar 2, 2025Updated last year
- ☆19Dec 4, 2025Updated 4 months ago
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆989Feb 5, 2026Updated 2 months ago
- Efficient triton implementation of Native Sparse Attention.☆275May 23, 2025Updated 11 months ago
- DeeperGEMM: crazy optimized version☆86May 5, 2025Updated 11 months ago
- 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.
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- Benchmark tests supporting the TiledCUDA library.☆18Nov 19, 2024Updated last year
- Fast and memory-efficient exact attention☆21Apr 10, 2026Updated 2 weeks ago
- Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning☆31Sep 12, 2025Updated 7 months ago
- ☆139May 29, 2025Updated 11 months ago
- a simple API to use CUPTI☆10Aug 19, 2025Updated 8 months ago
- [ICLR 2025 & COLM 2025] Official PyTorch implementation of the Forgetting Transformer and Adaptive Computation Pruning☆150Feb 25, 2026Updated 2 months ago
- Flash-Muon: An Efficient Implementation of Muon Optimizer☆248Jun 15, 2025Updated 10 months ago
- ☆279Jun 6, 2025Updated 10 months ago
- 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.
- A method for evaluating the high-level coherence of machine-generated texts. Identifies high-level coherence issues in transformer-based …☆11Mar 18, 2023Updated 3 years ago
- Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.☆478Updated this week
- Code for the paper "Function-Space Learning Rates"☆25Jun 3, 2025Updated 10 months ago
- ☆38Aug 7, 2025Updated 8 months ago
- ☆66Apr 26, 2025Updated last year
- mHC-lite: You Don’t Need 20 Sinkhorn-Knopp Iterations☆80Jan 12, 2026Updated 3 months ago
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆92Oct 30, 2024Updated last year
- NVIDIA cuTile learn☆167Dec 9, 2025Updated 4 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆273Oct 3, 2025Updated 6 months ago
- 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.
- Awesome Triton Resources☆40Apr 27, 2025Updated last year
- ☆70Jul 8, 2025Updated 9 months ago
- ☆22May 5, 2025Updated 11 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆345Updated this week
- Unofficial implementation of paper : Exploring the Space of Key-Value-Query Models with Intention☆12May 24, 2023Updated 2 years ago
- Medusa: Accelerating Serverless LLM Inference with Materialization [ASPLOS'25]☆44May 13, 2025Updated 11 months ago
- Code for the paper "Stack Attention: Improving the Ability of Transformers to Model Hierarchical Patterns"☆18Mar 15, 2024Updated 2 years ago
- Stick-breaking attention☆63Jul 1, 2025Updated 9 months ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆92Jul 17, 2025Updated 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.
- ☆11Oct 11, 2023Updated 2 years ago
- Transformers components but in Triton☆34May 9, 2025Updated 11 months ago
- H-Net Dynamic Hierarchical Architecture☆81Sep 11, 2025Updated 7 months ago
- Implementation from scratch in C of the Multi-head latent attention used in the Deepseek-v3 technical paper.☆18Jan 15, 2025Updated last year
- ☆24Jun 18, 2024Updated last year
- Checkpointable dataset utilities for foundation model training☆32Jan 29, 2024Updated 2 years ago
- Minimal PyTorch implementation of TP, SP, FSDP and sharded-EMA☆32Nov 27, 2025Updated 5 months ago