All 6 providers live

GPU Fleet Control
for AI Teams

Track, control, and budget GPU pods across 6 cloud providers — from CLI, mobile, or web. Stop runaway costs before they happen.

Already using podmon? Log in to dashboard

Quick Start

# Install

npm install -g podmon

# Authenticate

podmon auth login

# Connect a provider

podmon provider add runpod --api-key rpa_...

# Launch a pod

podmon create --gpu RTX_4090 --image pytorch/pytorch:latest

# Compare prices across all providers

podmon prices --gpu H100_SXM

Features

Everything you need to manage GPU workloads

6 Providers, 1 CLI

Manage pods across Prime Intellect, RunPod, Vast.ai, TensorDock, Lambda Labs, and Nebius from a single interface.

Real-Time Cost Tracking

Live burn rate, per-provider breakdowns, agent budgets, and 14-day spend trends — updated every second.

Agent Sandboxing

Assign budgets and GPU limits per autonomous agent. Claude Code, Codex, Aider — each gets its own guardrails.

Policy Gates

Block or warn on cost overruns, lifetime limits, restricted GPUs, time windows, and more. Rules enforced at creation time.

Mobile Kill Switch

Monitor pods and kill runaway workloads from your phone. Push notifications for budget alerts and pod failures.

Smart Provider Selection

Create a pod with just a GPU type — podmon finds the cheapest available provider and falls back automatically.

Providers

One CLI, six providers

All providers fully integrated with pod CRUD, GPU normalization, cost tracking, and availability queries.

Prime Intellect Live
RunPod Live
Vast.ai Live
TensorDock Live
Lambda Labs Live
Nebius Live
How It Works

Three steps to fleet control

1

Install & Connect

Install the CLI and add your provider API keys. Credentials never leave your machine.

2

Launch & Monitor

Create pods, set policies, and let the daemon track costs and health in real-time.

3

Control from Anywhere

Web dashboard, mobile app, or CLI. Kill runaway pods, check costs, and manage agents.