plugyawn
Follow
-
xmanager Public
Forked from google-deepmind/xmanagerA platform for managing machine learning experiments
Python Apache License 2.0 UpdatedOct 23, 2026 -
nextlat-reference-handoff Public
Focused historical NextLat joint-post-training reference for collaborator code review; no weights or original history
Python UpdatedOct 18, 2026 -
optimus Public
Research on zeroth-order training for transformers at scale.
-
modded-nanogpt Public
Forked from KellerJordan/modded-nanogptPython MIT License UpdatedOct 6, 2026 -
xls Public
Forked from google/xlsXLS: Accelerated HW Synthesis
C++ Apache License 2.0 UpdatedSep 12, 2026 -
tvm Public
Forked from apache/tvmOpen Machine Learning Compiler Framework
Python Apache License 2.0 UpdatedAug 11, 2026 -
-
diffusers Public
Forked from huggingface/diffusers🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
Python Apache License 2.0 UpdatedAug 1, 2026 -
pear Public
Pear workspace coordination repo — pins the Armada server and terminal submodules
Python UpdatedJul 30, 2026 -
armada-terminal Public
Armada terminal client — forked from zulip/zulip-terminal
Python Apache License 2.0 UpdatedJul 30, 2026 -
opentitan Public
Forked from lowRISC/opentitanOpenTitan: Open source silicon root of trust
SystemVerilog Apache License 2.0 UpdatedJul 27, 2026 -
triton-ext Public
Forked from triton-lang/triton-extA collection of out-of-tree extensions for the Triton language and compiler
Python Other UpdatedJul 26, 2026 -
TransformerEngine Public
Forked from NVIDIA/TransformerEngineA library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…
-
Megaprop Public
Newton-Muon + Preconditioned Optimizers for MoE Training at scale, with out-of-the-box support for MuP and FSDP support for Muon, built on top of Megatron-LM and TransformerEngine.
-
Megatron-LM Public
Forked from NVIDIA/Megatron-LMOngoing research training transformer models at scale
Python Other UpdatedJun 16, 2026 -
Emerging-Optimizers Public
Forked from NVIDIA-NeMo/Emerging-OptimizersPython Apache License 2.0 UpdatedJun 15, 2026 -
tokensieve Public
Save VLM prefill for VQA tasks by letting a policy select which tokens are important at test-time!
-
-
FlashOpt Public
Fast runtime for Random Optimization at scale.
-
-
opengenie Public
An open attempt at reproducing a simplified version of Google's Genie World Model API on a single H100.
-
kernels Public
Forked from huggingface/kernelsBuild compute kernels and load them from the Hub.
Python Apache License 2.0 UpdatedJun 4, 2026 -
lerobot Public
Forked from huggingface/lerobot🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning
Python Apache License 2.0 UpdatedMay 20, 2026 -
RL Public
Forked from NVIDIA-NeMo/RLScalable toolkit for efficient model reinforcement
-
tinygrad Public
Forked from tinygrad/tinygradYou like pytorch? You like micrograd? You love tinygrad! ❤️
-
-
nanodelight Public
Forked from karpathy/nanochatThe best ChatGPT that $150 can buy, now with delightful reinforcement.
Python MIT License UpdatedApr 21, 2026 -
-
awesome-physical-simulation Public
Forked from tsubasakong/awesome-nvidia-physical-simulationA curated list of resources for NVIDIA's physical-simulation stack — OpenUSD, Omniverse, Warp, PhysX, Newton, Isaac Sim, DSX, and more. Organized by architectural layer.
UpdatedApr 12, 2026 -