Documentation

Developer Docs

Everything you need to deploy and scale AI models on packet.ai's GPU cloud.

OpenAI-Compatible API

Use existing OpenAI SDKs and tools with your deployed models. Drop-in replacement with automatic model discovery, streaming support, and full compatibility.

Read the docs

Getting Started

Start deploying GPU workloads in minutes

Quick Start Guide

Deploy your first GPU in under 60 seconds

CLI Documentation

GPU cloud from your terminal

SSH Access

Connect to your GPU instances via SSH

Hugging Face Deployment

One-click model deployment from Hugging Face

API & Integration

Integrate with your applications

Bare Metal GPU API

Deploy and manage bare metal GPUs via API

REST API Reference

Complete API docs with Swagger UI

OpenAI-Compatible API

Drop-in replacement for OpenAI APIs

Token Factory

Managed inference with batch processing & LoRA

Inference Playground

Interactive chat interface for testing

Browser IDEs

VS Code and Jupyter in your browser

Monitoring & Analytics

Track performance and usage

GPU Metrics Dashboard

Real-time GPU monitoring and analytics

Token Usage

Track and analyze token consumption

Resources & Billing

Manage resources and costs

Pro 6000 Blackwell Models

Optimized models for 96GB GPUs

Persistent Storage

Attach and manage storage volumes

Persistent Workspace

Persist your environment across restarts

Budget Controls

Set spending limits and auto-shutdown

White-Label

Launch your own GPU cloud brand

Service Provider Guide

Set up your white-label GPU platform

Admin Guide

Internal guide for managing tenants

Need Help?

Can't find what you're looking for? Our team is here to help.

Contact Support support@packet.ai