Starred repositories
Open Source Continuous Inference Benchmarking Qwen3.5, DeepSeek, GPTOSS - GB200 NVL72 vs MI355X vs B200 vs GB300 NVL72 vs H100 & soon™ TPUv6e/v7/Trainium2/3
A pure-Python implementation of the Nvidia CuTe layout algebra intended to be approachable and easy to learn.
Let AI agents run research experiments + train small language models (in your browser!)
OpenClaw-RL: Train any agent simply by talking
A modular and open-ended toolkit for WebGPU, with advanced type inference and the ability to write shaders in TypeScript
Vite+ is the unified toolchain and entry point for web development. It manages your runtime, package manager, and frontend toolchain in one place.
A Chaos Engineering Platform for Kubernetes.
Engine-agnostic LLM gateway in Rust. Full OpenAI & Anthropic API compatibility across SGLang, vLLM, TRT-LLM, OpenAI, Gemini & more. Industry-first gRPC pipeline, KV cache-aware routing, chat histor…
Distributed pull-through cache for object storage with Kubernetes support
A 5-20x faster experimental Homebrew alternative
A completely vibecoded operating system targeting Aarch64
Algorithm powering the For You feed on X
GPU-accelerated line rendering for WebGPU with joins and caps
Interactive visualizations of the geometric intuition behind diffusion models.
A Virtual Machine Monitor for modern Cloud workloads. Features include CPU, memory and device hotplug, support for running Windows and Linux guests, device offload with vhost-user and a minimal com…
Accelerating MoE with IO and Tile-aware Optimizations
Development repository for the Triton language and compiler
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that sho…
Github mirror of trition-lang/triton repo.
A Vite plugin to replace imports for Lucide icons
Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.