Stars
cuda-oxide is an experimental Rust-to-CUDA compiler that lets you write (SIMT) GPU kernels in safe(ish), idiomatic Rust. It compiles standard Rust code directly to PTX — no DSLs, no foreign languag…
slotd is a lightweight, Rust-based job scheduler inspired by Slurm, designed for single-node, single-user workloads.
Low-latency AI engine for mobile devices & wearables
Academic Research Skills for Claude Code: research → write → review → revise → finalize
A minimal PyTorch re-implementation of Qwen 3.5
A GPU-oriented coverage-guided fuzzer for userland CUDA applications
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Qwen3.6 is the large language model series developed by Qwen team, Alibaba Group.
ANADAPTIVEEDGE-GUIDEDDUAL-NETWORKFRAMEWORKFORFASTQRCODE MOTIONDEBLURRING
A character-level language diffusion model trained on Tiny Shakespeare
WeDLM: The fastest diffusion language model with standard causal attention and native KV cache compatibility, delivering real speedups over vLLM-optimized baselines.
A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.
MobileFineTuner: Native C++ framework for fine-tuning LLMs directly on mobile devices. Features: LoRA/Full-FT, ZeRO-inspired parameter sharding, energy-aware throttling, custom autograd engine. Kee…
[DEIMv2] Real Time Object Detection Meets DINOv3
Visualize machine learning models with Netron in VSCode
🐉 Making Rust a first-class language and ecosystem for GPU shaders 🚧
Predictions of the four corners of documents.
An on-premises, OCR-free unstructured data extraction, markdown conversion and benchmarking toolkit. (https://idp-leaderboard.org/)
A tool for deep image processing's dataset augmentation
A Rust library integrated with ONNXRuntime, providing a collection of Computer Vison and Vision-Language models such as YOLO, FastVLM, and more.