Nebius AI Cloud API

Nebius AI Cloud API

This repository contains the configuration, manifests and Helm charts for deploying and managing Nebius' Kubernetes applications.

Run Slurm in Kubernetes

Run Slurm in Kubernetes

A Model Context Protocol (MCP) Gateway & Registry. Serves as a central management point for tools, resources, and prompts that can be accessed by MCP-compatible LLM applications. Converts REST API …

An extendible framework for executing benchmarks and computational experiments at scale

Module, Model, and Tensor Serialization/Deserialization

Module, Model, and Tensor Serialization/Deserialization

Module, Model, and Tensor Serialization/Deserialization

Enterprise Kubernetes management, accelerated. 🚀

[ARCHIVED] A university quantum algorithms/computation course supplement based on Qiskit

Qiskit is an open-source SDK for working with quantum computers at the level of extended quantum circuits, operators, and primitives.

The Ultimate all-in social media automation (outreach) tool 🤖

Granite Snack Cookbook -- easily consumable recipes (python notebooks) that showcase the capabilities of the Granite models

Production-leaning GPU Kubernetes baseline: NVIDIA device plugin, Node Feature Discovery, DCGM exporter, metrics, ingress, network policies, and safe defaults. Includes KIND for local smoke, Helm/K…

Repository for the WebXR Device API Specification.

Shell script toolkit to automate VIOS operations (vSCSI, NPIV, SEA) via HMC CLI and viosvrcmd.

Terraform IaC templates for provisioning AIX and Linux LPARs in IBM Power Virtual Server.

Python CLI to query IBM HMC REST APIs for LPAR inventory/utilization and perform dynamic CPU/memory adjustments.

Minimal Slurm cluster for demos: docker-compose for slurmctld/slurmdbd/2×slurmd, plus Ansible playbooks for VM setup. Includes example CPU/GPU sbatch jobs, NCCL all-reduce test, Prometheus exporter…

NVIDIA Triton Inference Server on Kubernetes with a minimal model repo (e.g., ResNet50 ONNX), Helm chart, HPA, and Locust load test. Includes Prometheus/Grafana metrics and a ready-to-import dashbo…

Practical MLOps template: Argo Workflows pipeline (prep → train → evaluate → register) tracked in MLflow, then deployed as a FastAPI service. Includes Dockerfiles, K8s manifests, HPA, Makefile, and…

Turn-key PyTorch DistributedDataParallel (DDP) with NCCL for single- and multi-node GPU training. Includes Docker image, K8s Job and optional Kubeflow PyTorchJob, Slurm sbatch, sensible NCCL env de…

IBM Sytem i AS400 Projects

Generate, animate and schedule your AI characters 🤖