Official codebase for the MLSys 2026 paper "IntAttention: A Fully Integer Attention Pipeline for Efficient Edge Inference". It enables high-fidelity and high-speed LLM/ViT deployment on ARM CPUs.
☆23May 29, 2026Updated 5 months ago
Alternatives and similar repositories for IntAttention
Users that are interested in IntAttention are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆20Oct 22, 2026Updated last week
- Top-level repository of ACT Ecosystem☆36Oct 20, 2026Updated last week
- ☆19Mar 29, 2026Updated 7 months ago
- ☆20Apr 26, 2026Updated 6 months ago
- [MLSys 26] 🥇 Solution for Gated Delta Net Track of MLSys 26 Flash infer competition☆36May 22, 2026Updated 5 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.
- Official repository for paper Routing-Free Mixture-of-Experts.☆19Aug 23, 2026Updated 2 months ago
- [ICCD26] A fast simulator for mixed-precision RRAM and SRAM compute-in-memory AI processors☆12Dec 11, 2025Updated 10 months ago
- CUDA MoE kernels.☆12Sep 21, 2024Updated 2 years ago
- A multilayer perceptron (for simple image classification), accelerated with CUDA☆17Oct 21, 2019Updated 7 years ago
- ☆18Jun 26, 2024Updated 2 years ago
- Your Ryzen AI Max is running at half speed. SnowLLM: up to 9x faster prefill and 2x faster decode than llama.cpp on the same GGUF. Qwen3.…☆13Oct 22, 2026Updated last week
- ☆15Jul 25, 2025Updated last year
- ☆16Sep 25, 2025Updated last year
- Tempo is a system for declarative, efficient, end-to-end compiled dynamic deep learning☆31Oct 21, 2025Updated last year
- 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.
- DeDe (OSDI '25): an optimization framework for large-scale resource allocation☆15May 18, 2026Updated 5 months ago
- ☆64Jul 21, 2024Updated 2 years ago
- momoten is an experimental translation and runtime layer that implements a restricted, buffer-oriented subset of the Vulkan compute API o…☆27Updated this week
- writing really fast kernels☆21Oct 11, 2026Updated 2 weeks ago
- An unbiased CPU benchmark by OpenCV that provides an evaluation of different CPUs under real-world computer vision and AI workloads.☆12Aug 20, 2026Updated 2 months ago
- ☆43Oct 12, 2025Updated last year
- ☆19Jan 8, 2026Updated 9 months ago
- Fast Emulation of Approximate ViT models in PyTorch☆13Feb 13, 2024Updated 2 years ago
- serving a torch model using Celery, Redis and RabbitMQ to serve users asynchronously☆25Jan 21, 2024Updated 2 years ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- Dist-MoE implements Speed-of-Light distributed Mixture-of-Experts forward and backward kernels. They are highly optimized and fused comm-…☆62Updated this week
- Quantized Training for Convolutional Neural Networks using Xilinx Brevitas☆12Mar 16, 2022Updated 4 years ago
- Framework for Algorithmic Correctness Testing of Operators☆18Mar 9, 2026Updated 7 months ago
- Algorithm-System Co-design: accurate and efficient 2-bit KV cache quantization for LLM Inference.☆23May 20, 2026Updated 5 months ago
- A high-performance, RDMA-native data-plane library for large-scale heterogeneous memory pools, developed by the Tencent Astral Network Te…☆35Aug 6, 2026Updated 2 months ago
- Flash-Flash KDA: H100-optimized Flash Kimi Delta Attention kernels☆24Jul 24, 2026Updated 3 months ago
- A CUDA kernel for NHWC GroupNorm for PyTorch☆26Nov 15, 2024Updated last year
- Batched square compact-Householder QR factorization.☆15Jul 2, 2026Updated 3 months ago
- VECBEE: A Versatile Efficiency-Accuracy Configurable Batch Error Estimation Method for Greedy Approximate Logic Synthesis☆13Mar 8, 2022Updated 4 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.
- Benchmarks for Approximate Circuit Synthesis☆17Aug 2, 2020Updated 6 years ago
- 面向开发者与初学者的 nano-vLLM 交互式源码教程 - 通过 13 个 HTML 互动实验 + 13 章中文教程理解大模型推理引擎☆38Aug 23, 2026Updated 2 months ago
- ☆21Mar 3, 2025Updated last year
- A collection of graph contrastive learning methods.☆18Apr 1, 2022Updated 4 years ago
- ☆19Dec 21, 2020Updated 5 years ago
- ☆18Jul 31, 2023Updated 3 years ago
- ☆74Mar 24, 2026Updated 7 months ago