PiotrNawrot / sparse-frontierLinks
The evaluation framework for training-free sparse attention in LLMs
☆114Updated this week
Alternatives and similar repositories for sparse-frontier
Users that are interested in sparse-frontier are comparing it to the libraries listed below
Sorting:
- ☆132Updated 8 months ago
- An efficient implementation of the NSA (Native Sparse Attention) kernel☆128Updated 7 months ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆92Updated 6 months ago
- Flash-Muon: An Efficient Implementation of Muon Optimizer☆229Updated 7 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆131Updated last year
- [ICML 2025] From Low Rank Gradient Subspace Stabilization to Low-Rank Weights: Observations, Theories and Applications☆52Updated 3 months ago
- Fast and memory-efficient exact attention☆75Updated 10 months ago
- Kinetics: Rethinking Test-Time Scaling Laws☆85Updated 6 months ago
- ☆158Updated 11 months ago
- ☆63Updated 7 months ago
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆60Updated last year
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆52Updated last year
- ☆104Updated 11 months ago
- Official implementation for Training LLMs with MXFP4☆118Updated 9 months ago
- ☆85Updated 2 months ago
- The official repository for SkyLadder: Better and Faster Pretraining via Context Window Scheduling☆42Updated last month
- Work in progress.☆79Updated 2 months ago
- ☆83Updated 2 years ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆93Updated last year
- 🔥 LLM-powered GPU kernel synthesis: Train models to convert PyTorch ops into optimized Triton kernels via SFT+RL. Multi-turn compilation…☆116Updated 2 months ago
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆51Updated 6 months ago
- ☆54Updated last year
- This repository contains code for the MicroAdam paper.☆22Updated last year
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆148Updated last year
- ☆15Updated last year
- Code for studying the super weight in LLM☆120Updated last year
- ☆66Updated 10 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆137Updated last year
- Muon fsdp 2☆52Updated 5 months ago
- Repository for the Q-Filters method (https://arxiv.org/pdf/2503.02812)☆35Updated 10 months ago