Implemented and benchmarked KV cache compression methods for LLM inference in Triton. Optimized kernels for KIVI and TurboQuant with Flash Attention integration - up to 7.11× compression on GPT-2 and 7.31× on Qwen3.5 9B.
☆18Jun 19, 2026Updated 4 months ago
Alternatives and similar repositories for KV-Compression
Users that are interested in KV-Compression are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- A CUDA Python experiment demonstrating kernel fusion by combining ReLU and LayerNorm into a single GPU pass and comparing it against the …☆14Feb 28, 2026Updated 8 months ago
- Fastest du replacement and file tree analysis for 100+GB, million-file worktrees☆18Updated this week
- Flash Attention from scratch, tiled CUDA forward kernel, online softmax with running max and correction factor, recomputation trick in ba…☆19Mar 6, 2026Updated 8 months ago
- ☆24Jul 20, 2026Updated 3 months ago
- everything you need to know about inference engineering☆45Sep 29, 2026Updated last month
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- a minimal paged attention implementation☆20Jan 30, 2026Updated 9 months ago
- A fast image processing TUI written in Rust☆22Aug 18, 2025Updated last year
- Learning notes for understanding modern Machine Learning System.☆66Updated this week
- a collection of CUDA Kernels☆18Jul 24, 2026Updated 3 months ago
- Set of tools to parse DS2 morpheme binaries, and to compile them☆17Oct 25, 2026Updated 2 weeks ago
- Speed-of-Light SW efficiency by using ultra low-latency primitives for comms collectives☆13Apr 8, 2026Updated 7 months ago
- Aspiring universal programming language. Features: user-definable syntax and types + self-improving optimizer that creates provably corre…☆23Updated this week
- Djinnlang implementation for Rust☆27Oct 25, 2026Updated 2 weeks ago
- Conference on Homotopy Type Theory 2019☆16Sep 18, 2019Updated 7 years 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.
- Fused SwiGLU Triton kernels☆15Jan 25, 2024Updated 2 years ago
- Hacks for PyTorch☆19Apr 18, 2023Updated 3 years ago
- Reconstruction of the Dimuon optimizer☆15Oct 28, 2026Updated 2 weeks ago
- Interactive 3D model of NVIDIA Vera Rubin hardware (superchip, NVL72 compute tray, HGX NVL8) with connection tracing and a guided tour☆17Oct 1, 2026Updated last month
- AC No Code 是偷懒者最好的在OJ中写代码AC的方式: Write nothing; submit nowhere.☆10May 18, 2020Updated 6 years ago
- gen-dsp converts code exported from a Max gen~ object and a graph-oriented dsl into multiple dsp plugin architectures.☆23Nov 4, 2026Updated last week
- The IUCr CIF core dictionary☆24Oct 20, 2026Updated 3 weeks ago
- A proof that Z is Diophantine in Q.☆67Updated this week
- A notebook using numpy, scipy, and matplotlib to show the concepts taught in the course Linear Algebra by Prof. Gilbert Strang at MIT☆11Jul 16, 2022Updated 4 years ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- ☆14Aug 7, 2026Updated 3 months ago
- Plug-and-play PyTorch transformer research framework. Swap attention, FFN, norm, positional, connection, projection, optimizer, scheduler…☆18Oct 25, 2026Updated 2 weeks ago
- ☆15Sep 29, 2026Updated last month
- TopicDeck is a column based topical news reader☆12Mar 4, 2023Updated 3 years ago
- mKernel: fast multi-node, multi-GPU fused kernels☆309Oct 17, 2026Updated 3 weeks ago
- Simply-typed lambda calculus (plus several features) -> x86 executables for Linux and Windows☆25Jan 25, 2013Updated 13 years ago
- Sparsity as a Variance Regulator for Improved Depth Utilization in Language Models☆16May 11, 2026Updated 6 months ago
- BFloat16 Fused Adam Operator for PyTorch☆20Nov 16, 2024Updated last year
- Authenticate, monitor and control your swarm of Folding@home clients remotely☆12Oct 1, 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.
- Lean 4 + the Sail-generated RISC-V ISA model prove, CompCert-style, that an inductive WHILE semantics abstracts a real interpreter binary…☆60Oct 5, 2026Updated last month
- ☆14Sep 12, 2021Updated 5 years ago
- Sources of demo for the presentation: Data Navigation with Pathom 3☆11Oct 26, 2021Updated 5 years ago
- [NeurIPS'26] Open-source native multimodal pretraining — without catastrophic forgetting.☆15Jul 2, 2026Updated 4 months ago
- ☆19Aug 24, 2022Updated 4 years ago
- Windows code for communicating with a VC707 evaluation board running a MIAOW compute unit.☆11Oct 20, 2015Updated 11 years ago
- fzf-based test selection with pytest☆16Jan 26, 2026Updated 9 months ago
Are these results useful?