Machine Learning Systems: Foundations, Scaling, Agentic AI, and Physical AI (Vols I–IV) • Harvard CS249r | https://mlsysbook.ai

machine-learning deep-learning robotics artificial-intelligence textbook courseware computer-systems ai-agents edge-machine-learning machine-learning-systems mlsys tinyml embodied-ai cloud-ml embedded-ml agentic-ai mobile-ml physical-ai

• Updated Nov 12, 2026
• Python

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

ai mlsys aiinfra dlsys aisys

• Updated Sep 3, 2025
• Jupyter Notebook

The RL Bridge for LLM-based Agent Applications. Made Simple & Flexible.

agent reinforcement-learning rl machine-learning-systems mlsys llm llm-agent llm-reasoning

• Updated Nov 12, 2026
• Python

🚀 Awesome System for Machine Learning ⚡️ AI System Papers and Industry Practice. ⚡️ System for Machine Learning, LLM (Large Language Model), GenAI (Generative AI). 🍻 OSDI, NSDI, SIGCOMM, SoCC, MLSys, etc. 🗃️ Llama3, Mistral, etc. 🧑‍💻 Video Tutorials.

model-serving model-training mlsys ai-infra large-language-models genai llmsys

• Updated Jul 25, 2025

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

cuda triton attention vit quantization video-generation mlsys inference-acceleration efficient-attention llm llm-infra video-generate

• Updated Jan 17, 2026
• Cuda

[ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

flux lora quantization iclr diffusion-models mlsys comfyui genai iclr2025

• Updated Sep 6, 2026
• Python

ComfyUI Plugin of Nunchaku

flux quantization diffusion mlsys comfyui genai

• Updated Sep 6, 2026
• Python

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

attention vit quantization video-generation mlsys inference-acceleration ai-infra vision-transformer sparse-attention llm sageattention

• Updated Feb 25, 2026
• Cuda

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

transformer video-generation mlsys inference-acceleration ai-infra linear-attention sparse-attention diffusion-transformer train-acceleration sparse-linear-attention

• Updated Oct 18, 2026
• Python

A model compilation solution for various hardware

tensorflow llvm pytorch onnx mlsys mlir llm

• Updated Aug 20, 2025
• MLIR

[ACL 2026] Towards Efficient Large Language Model Serving: A Survey on System-Aware KV Cache Optimization

machine-learning ai system computer-architecture neural-language-processing mlsys kv-cache serving-ml llm llm-serving llm-inference

• Updated Oct 23, 2026
• Python

FedScale is a scalable and extensible open-source federated learning (FL) platform.

benchmark machine-learning deep-learning deployment tensorflow pytorch distributed dataset osdi icml federated-learning mlsys

• Updated Dec 18, 2023
• Python

Measure and optimize the energy consumption of your AI applications!

energy deep-learning mlsys

• Updated Nov 4, 2026
• Python

The repository has collected a batch of noteworthy MLSys bloggers (Algorithms/Systems)

machine-learning machine-learning-systems mlsys llm llm-training llm-inference

• Updated Jan 5, 2025
• HTML

A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems

dataset mlsys llm llm-serving

• Updated Jun 30, 2026
• Python

Machine Learning Framework for Operating Systems - Brings ML to Linux kernel

machine-learning kernel-module linux-kernel operating-systems auto-tuning mlsys

• Updated Dec 13, 2021
• C

An acceleration library that supports arbitrary bit-width combinatorial quantization operations

research cuda mlsys quantized-networks llm-inference

• Updated Sep 30, 2024
• C++

A scalable & efficient active learning/data selection system for everyone.

machine-learning deep-learning pytorch active-learning automl mlops mlsys

• Updated Jul 8, 2024
• Python

An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682

machine-learning gpu cuda gpu-programming kernel-fusion mlsys llm-inference agent-harness megakernel

• Updated Sep 18, 2026
• Python

Wave Forcing: Towards Speed-of-Light Streaming Video Generation

streaming-video video-generation mlsys efficient-ai

• Updated Nov 5, 2026
• Python