a personal collection of my notes for ml sys
☆124Sep 1, 2026Updated last month
Alternatives and similar repositories for ml-systems-notes
Users that are interested in ml-systems-notes 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 LLM inference engine to run on consumer hardware☆48Apr 15, 2026Updated 5 months ago
- writing really fast kernels☆20Sep 12, 2026Updated 3 weeks ago
- A GPT-2 inference engine written from scratch in CUDA and C++. Implements custom CUDA kernels for tiled matrix multiplication, LayerNorm,…☆44May 17, 2026Updated 4 months ago
- fastcv is a CUDA rewrite of the opencv filters with python bindings☆75Oct 10, 2025Updated 11 months ago
- a minimal paged attention implementation☆20Jan 30, 2026Updated 8 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Repository for the CUDA H100 Course☆76Apr 12, 2026Updated 5 months ago
- A Python DSL for Apple Metal GPU compute☆26Jul 23, 2026Updated 2 months ago
- ☆31Sep 16, 2023Updated 3 years ago
- ☆91Dec 16, 2025Updated 9 months ago
- Code companion for the RL Post-Training Handbook - training reasoning models on a single GPU☆20Jan 30, 2026Updated 8 months ago
- Focused on a consistent API across multiple Agent Harnesses like Claude Code, Langchain Deep Agents, Letta, Goose and more.☆24Jan 23, 2026Updated 8 months ago
- ☆51Aug 17, 2026Updated last month
- This Repo Contains all the materials of Pattern Recognition and Neural Networks course done at IISc and recorded for NPTEL - Mathematical…☆33Mar 17, 2026Updated 6 months ago
- minimal pytorch 4D parallelism☆73Feb 16, 2026Updated 7 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- An educational distributed training and inference library for neural nets using local computing☆86Jun 10, 2026Updated 3 months ago
- 《汇编语言一发入魂》配套代码☆15May 30, 2020Updated 6 years ago
- A PyTorch implementation of the GPT-OSS-20B architecture. All components are coded from scratch: RoPE with YaRN, RMSNorm, SwiGLU with cla…☆237Dec 2, 2025Updated 10 months ago
- NCCL communication API layer, and transport layer created from first principles.☆16Aug 20, 2025Updated last year
- A series of high-performance GEMM (General Matrix Multiply) implementations Iteratively optimised for H100 GPUs in Pure CUDA.☆83Feb 18, 2026Updated 7 months ago
- A curated resource list for learning AI performance engineering, from GPU fundamentals to production inference.☆3,946Sep 12, 2026Updated 3 weeks ago
- Matrix multiplication on GPUs for matrices stored on a CPU. Similar to cublasXt, but ported to both NVIDIA and AMD GPUs.☆33Apr 2, 2025Updated last year
- Stas' Python Cookbook - Python recipes that I use daily☆65Jul 2, 2026Updated 3 months ago
- Curated collection of AI inference engineering resources — LLM serving, GPU kernels, quantization, distributed inference, and production …☆326Feb 4, 2026Updated 7 months ago
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- Row-wise block scaling for fp8 quantization matrix multiplication. Solution to GPU mode AMD challenge.☆19Feb 9, 2026Updated 7 months ago
- ✨ AI/ML Research☆457Updated this week
- ☆91Updated this week
- ☆22Jan 2, 2026Updated 9 months ago
- a simple c++ inference engine for gpt based architecture☆41Dec 10, 2025Updated 9 months ago
- Quantized LLM training in pure CUDA/C++.☆261Aug 31, 2026Updated last month
- Write a fast kernel and see how you compare against the best humans and AI on gpumode.com☆114Sep 18, 2026Updated 2 weeks ago
- High-Performance FP32 GEMM on CUDA devices☆130Jan 21, 2025Updated last year
- ☆19Mar 29, 2026Updated 6 months 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.
- IMMA-based **FP8-as-storage** GEMM experiments for Ampere (sm_86 / RTX 3090 Ti).☆27Jan 30, 2026Updated 8 months ago
- ☆27Aug 24, 2026Updated last month
- ☆15Mar 11, 2026Updated 6 months ago
- Triton kernels and PyTorch ops for Block Attention Residuals (AttnRes)☆89May 29, 2026Updated 4 months ago
- Build a Content-Based Movie Recommender System (TF-IDF, BM25, BERT)☆15Jun 13, 2022Updated 4 years ago
- Rust based Unix like OS, written by AI, guided by Humans.☆15Dec 8, 2025Updated 9 months ago
- A dynamic binary instrumentation tool for tracing and analyzing GPU kernel instructions.☆80Updated this week