Dev
GitHub repos gaining traction - what high-signal users are starring and what's climbing the board, captured daily and enriched from GitHub. Raw material for spotting new tech and patterns worth building on.
3,195
repos tracked
105
surfaced this week
193
created < 30d
Python
top language
42 repos
-
reproducing the flapping airplanes megakernel, in collab with magni
-
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
-
Application Kernel for Containers
-
A tiny stdin/stdout worker that keeps a Python process alive and returns concise text for each request
-
Experimental mini python jupyter kernel
-
Linux kernel source tree
-
kernelboard is the webapp for https://www.gpumode.com
-
NVIDIA's CUDA platform for Rust. Host runtime crates plus Tile (cutile-rs) and SIMT (cuda-oxide) kernel programming models in idiomatic Rust.
-
A PyTorch native automated mega-kernel compiler
-
Two case studies: megakernels roughly on par with CUDA Graphs + PDL. Code and baselines for further research.
-
Everything a Jupyter kernel needs except the language
-
A Quirky Assortment of CuTe Kernels
-
Automatic formal verification for warp-specialized GPU kernels.
-
Phi-Bench (Φ-Bench): 85 open-source LLM-infrastructure engineering tasks for frontier LLMs & coding agents (KFC/LH/E2E). Self-contained public Dockerfile + offline scoring + reference solution per task. Website: llminfrabench.com
-
A native desktop IDE for Lean 4 on macOS, Windows and Linux. Every proof re-checked by Tenet, an independent Lean 4 kernel. An MCP server for AI assistants, with Git and GitHub built in.
-
A collection of megakernels for TPU
-
Minimal Jupyter protocol commons: wire-format sessions and kernel-client call conventions
-
Curated high-performance TIRx GPU kernels
-
Kernel Design Agents (KDA) is a agent-centric workflow to write high-performance CUDA Kernels.
-
Pyskills for Answer.AI coding sessions: skill text and companion tooling for clikernel
-
Concurrent-safe Jupyter KernelClient
-
Official Problem Sets / Reference Kernels for the GPU MODE Leaderboard!
-
Winner 🏆 (Agent-only) MLSys 2026 - FlashInfer AI Kernel Generation Contest for the DeepSeek Sparse Attention (DSA) track with an average speedup of 34.93x
-
The J language in notebooks, with j magics for IPython and Jupyter and a kernmini-based J kernel
-
FlashInfer: Kernel Library for LLM Serving
-
DeepGEMM: clean and efficient BLAS kernel library on GPU
-
Tile primitives for speedy kernels
-
Websocket gateway for Jupyter kernels
-
Helpers for ipykernel and friends
-
Fused MoE kernel
-
Minimal ipykernel subclass that removes a troublesome lock
-
A persistent, real Jupyter kernel for agents.
-
Write a fast kernel and see how you compare against the best humans and AI on gpumode.com
-
Apple Silicon (Metal) backend for Triton: write standard @triton.jit kernels on your Mac GPU. The same source runs bit-identical on NVIDIA and AMD, so you develop kernel logic locally and rent a datacenter GPU only for the perf pass.
-
high-performance linear attention kernel library built on TileLang
-
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
-
Production-Grade Autoresearch. Ideal for agent harness engineering, prompt engineering, ML model development, GPU kernels, and other optimizable code.
-
Triton kernels for dynamic causal short convolutions.