Vercel for Edge AI

Deploy AI models to any edge device.
One command.

Fleet management, staged rollouts, and real-time monitoring for production edge AI deployments.

Terminal
$
Capabilities

Everything you need for
production edge AI

Deploy

One-command deployment across your fleet

No SSH, no manual file transfers, no dependency hell. One CLI command optimizes, distributes, and verifies your model across every device in your fleet.

Fleet

See every device at a glance

Track which models are deployed where, catch failures before they impact users, and manage your entire edge infrastructure from a single pane of glass.

Staged Rollouts

Deploy to 10% first, then advance

Canary your deployments, validate performance against baselines, and roll back automatically if something goes wrong. No more 3am incidents.

Monitoring

CPU, memory, GPU, latency, error rates in real time

Spot hardware degradation, model performance issues, and capacity problems before they become outages. Every metric, every device, always.

Getting Started

Three commands.
Zero to deployed.

01

Install the CLI

Install the EdgeKiwi CLI from crates.io. Works on macOS, Linux, and Windows.

$cargo install kiwi-cli
02

Register your devices

Start the agent on each edge device. It connects to your fleet automatically.

$kiwi agent start --api-key <YOUR_KEY>
03

Deploy your model

Initialize a project with your model and deploy to your entire fleet in one step.

$kiwi init --model llama3.2-3b && kiwi deploy
Why EdgeKiwi

Built for production edge AI

Built in Rust

Memory-safe, zero-cost abstractions, and blazing fast. No garbage collection pauses on your edge devices.

Open Source Core

CLI, agent, and core library are Apache 2.0. Inspect the code, contribute, and build with confidence.

GPU Accelerated

Native support for CUDA, Metal, and Vulkan. Automatically optimizes for your device's hardware.

Enterprise Ready

Fleet management, staged rollouts, rollback, and real-time monitoring for production deployments.

Pricing

Simple, transparent pricing

All plans include a 14-day free trial. No credit card required.

Starter

$29/mo

For small teams getting started with edge AI.

  • 5 devices
  • Full CLI access
  • Model optimization
  • Community support
  • 14-day free trial
Recommended

Team

$99/mo

For teams managing production edge deployments.

  • 20 devices
  • Staged rollouts
  • Fleet monitoring
  • A/B testing
  • Email support
  • 14-day free trial

Enterprise

Custom

For organizations with large-scale edge fleets.

  • Unlimited devices
  • Custom SLAs
  • Priority support
  • Dedicated account manager
  • Custom integrations
  • SSO / SAML