Deployment Overview

#Instill Cloud

Try Instill Cloud for the easiest and most reliable experience to access all the fantastic features of Instill Core.

#Self-host

You have the option to self-host locally using Instill CLI, Docker Compose, Kubernetes with Helm, and more. We are continually optimizing the self-hosting process to get you up and running with our services in just a few minutes. If you encounter any challenges with self-deployment, please don't hesitate to reach out to us on GitHub or join our Discord community for assistance.

#Docker Resource Requirements

To successfully launch Instill Core, it's essential to meet the following Docker Engine resource requirements. Typically, launch failures occur due to inadequate virtual disk size. We strongly recommend configuring the Docker Engine with the following settings for optimal performance:

In order to launch Instill Core successfully, it's imperative to ensure that your Docker Engine meets the following resource requirements. Often, launch failures are attributed to insufficient virtual disk space. To achieve optimal performance, we strongly recommend configuring your Docker Engine with the following settings:

Docker Engine ResourceMinimal RequirementsRecommended Requirements
CPU Limit28
Memory Limit2 GB32 GB
Swap1 GB1 GB
Virtual Disk Limit64 GB256 GB

Please note that the initial launch process may take up to 1 hour, contingent on your internet speed. The most time-consuming aspect of the launch is the download of the Triton Inference Server image for Instill Model, which is approximately 11 GB in size. Subsequent launches will be considerably quicker, usually completing in under 5 minutes.

#Deploy with Instill CLI

Please checkout Instill CLI. This is the recommended way to access both Instill Core and Instill Cloud. For the best experience accessing both Instill Core and Instill Cloud, we recommend using the Instill CLI.

#Deploy with other methods

Alternatively, you can deploy Instill Core via:

Last updated: 1/15/2024, 2:25:21 PM