GLM-ASR-Nano: A robust, open-source speech recognition model with 1.5B parameters

NVSentinel is a cross-platform fault remediation service designed to rapidly remediate runtime node-level issues in GPU-accelerated computing environments

Checkpoint-engine is a simple middleware to update model weights in LLM inference engines

Nsight Python is a Python kernel profiling interface based on NVIDIA Nsight Tools

Autonomous GPU Kernel Generation via Deep Agents

Primus-SaFE(Stability and Fault Endurance)

Minimalistic 4D-parallelism distributed training framework for education purpose

An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

Distribute and run AI workloads on Kubernetes magically in Python, like PyTorch for ML infra.

Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond

agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes.

The best ChatGPT that $100 can buy.

Beads - A memory upgrade for your coding agent

Claude Code superpowers: core skills library

Open Source Continuous Inference Benchmarking - GB200 NVL72 vs MI355X vs B200 vs H200 vs MI325X & soon™ TPUv6e/v7/Trainium2/3/GB300 NVL72 - DeepSeek 670B MoE, GPTOSS

A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

GenAI inference performance benchmarking tool

A storage solution for PyTorch tensors with distributed tensor support.

PyTorch-native post-training at scale

Post-training with Tinker

GitHub Copilot CLI brings the power of Copilot coding agent directly to your terminal.

nv-one-logger enables tracking of GPU application progress over time and can help to identify overhead from workload and cluster inefficiencies to provide efficiency metrics.

Nano vLLM

Agent Builder and Runtime by Docker Engineering

Intelligent Router for Mixture-of-Models

Environments for LLM Reinforcement Learning

Pipeline parallelism for the minimalist