Lists (1)
Sort Name ascending (A-Z)
Stars
Debugging tool to print information (especially sharding) about jax arrays
WFGY 2.0. Semantic Reasoning Engine for LLMs (MIT). Fixes RAG/OCR drift, collapse & “ghost matches” via symbolic overlays + logic patches. Autoboot; OneLine & Flagship. ⭐ Star if you explore semant…
Everything you want to know about Google Cloud TPU
Official PyTorch implementation and models for paper "Diffusion Beats Autoregressive in Data-Constrained Settings". We find diffusion models are significantly more data-efficient than standard left…
[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without lossing end-to-end metrics across language, image, and video models.
https://wavespeed.ai/ Context parallel attention that accelerates DiT model inference with dynamic caching
[NeurIPS 2024] Simple and Effective Masked Diffusion Language Model
NanoGPT speedrun in JAX. Originally at https://nor-git.pages.dev/modded-nanogpt-jax/
Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.
Reference PyTorch implementation and models for DINOv3
The Institutional Data Initiative's pipeline for analyzing, refining, and publishing the Institutional Books 1.0 collection.
An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.
Environments for LLM Reinforcement Learning
A fast and efficient type assistant for Python, including tensor shape inference
Ever used asyncio and wished you hadn't? A tiny (~300 lines) event loop for Python.
EXO Gym is an open-source Python toolkit that facilitates distributed AI research.
main-horse / hnet-old
Forked from goombalab/hnetH-Net Dynamic Hierarchical Architecture
Official PyTorch implementation for "Effective and Efficient Masked Image Generation Models"
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…
Unbearably fast near-real-time pure-Python runtime-static type-checker.
A series of math-specific large language models of our Qwen2 series.
OWL Control is a desktop application that records gameplay footage and input data from video games to create open-source datasets for AI research.