← Home

Self-Hosting

Curated entries for self-hosting from the SMF Clearinghouse.

Showing 11 of 11

Hardware

Intel-Based Self-Hosting

Intel Arc GPU and Xeon CPU setups for OpenVINO, IPEX, and CPU-optimized inference.

IntelArcOpenVINO
Read
Virtualization

Kubernetes Self-Hosting

Run agent workloads, vector databases, and inference APIs on Kubernetes for scale, resilience, and reproducible deployments.

kubernetesk8scontainers
Verified 29 days ago
Read
Hardware

Mac-Based Self-Hosting

Apple Silicon local inference from entry-level MacBook Air to Mac Studio Ultra.

AppleMacMLX
Read
Operating System

macOS Self-Hosting

Apple Silicon native inference with MLX, llama.cpp Metal backends, and local agent tools.

macOSApple SiliconMLX
Read
Operating System

NixOS Self-Hosting

Reproducible, declarative AI infrastructure with NixOS — ideal for teams that want versioned system configurations and rollback safety.

nixosnixreproducible
Verified 29 days ago
Read
Hardware

Single-Board Edge Self-Hosting

Run small models and lightweight agents on Raspberry Pi, Orange Pi, and other edge boards for offline, low-power inference.

edgeraspberry-pisingle-board
Verified 29 days ago
Read
Virtualization

Proxmox VE Self-Hosting

Run AI workloads and agent VMs on Proxmox VE. Combine bare-metal performance with VM isolation, GPU passthrough, and easy snapshots.

proxmoxvirtualizationkvm
Verified 36 days ago
Read
Hardware

NVIDIA-Based Self-Hosting

The complete guide to CUDA-accelerated local AI — from DGX Spark and RTX workstations to A100/H100 datacenter clusters.

NVIDIACUDARTX
Verified 44 days ago
Read
Hardware

AMD-Based Self-Hosting

A practical guide to ROCm-accelerated local AI on AMD Radeon and Instinct hardware — from budget gaming GPUs to MI300X datacenter clusters.

AMDROCmRadeon
Verified 44 days ago
Read
Operating System

Linux Self-Hosting

The definitive OS for local AI. Ubuntu setup, drivers, containers, and the inference engines that power agent workloads.

LinuxUbuntuROCm
Verified 44 days ago
Read
Operating System

Windows Self-Hosting

Run local LLMs on Windows with WSL2, native CUDA, and tools like LM Studio and Ollama.

WindowsWSL2CUDA
Read