Skip to content
Tabby Coding Assistant AI Agents · Tabby v0.32.0 Launch on AWS

Getting started

This page takes you from the AWS Marketplace listing to a running Tabby server: subscribe, launch, connect, check that Tabby is healthy, and reach its API from your computer. It takes about ten minutes.

You need: an AWS account with permission to subscribe to AWS Marketplace products and launch EC2 instances. To reach the API from your computer, you also need the AWS CLI and the Session Manager plugin.

1. Subscribe

Open the Tabby Coding Assistant listing, review the pricing and terms, and subscribe.

2. Launch the instance

Launch from the listing with these settings:

  • Instance type: t3.medium, or larger if you plan to run a local model.
  • Storage: at least 16 GiB, encrypted gp3. More if you plan to download model weights.
  • Metadata: IMDSv2.
  • Security group: no inbound rules are needed. Tabby listens on loopback only. Allow outbound HTTPS.
  • IAM instance profile: attach a role with the AWS managed policy AmazonSSMManagedInstanceCore so you can connect with Session Manager. Tabby itself needs no AWS permissions.

3. Connect

Connect with Session Manager, from the EC2 console (Connect > Session Manager) or with the AWS CLI:

aws ssm start-session --target <instance-id>

Wait for first boot to finish:

cloud-init status --wait

4. Run the walkthrough

sudo -iu ubuntu bash /opt/sushi/docs/demo.sh

The walkthrough runs the self-test, then checks that tabby.service is active, that Tabby and the embedding stub listen on 127.0.0.1 only, that /v1/health answers, and that a completion request is refused because no model is attached yet. Each step prints STEP, COMMAND and RESULT lines, and the script ends with SUSHI_CUSTOMER_DEMO_PASSED.

Run it before you attach a model

The walkthrough and the self-test expect the default configuration, with no completion model. Run them first.

5. Check health yourself

curl http://127.0.0.1:8080/v1/health

The JSON reply reports "device":"cpu" and the Tabby version.

6. Reach the API from your computer

Because Tabby listens on loopback, reach it with Session Manager port forwarding. On your computer:

aws ssm start-session --target <instance-id> \
  --document-name AWS-StartPortForwardingSession \
  --parameters '{"portNumber":["8080"],"localPortNumber":["8080"]}'

Then, in another terminal:

curl http://127.0.0.1:8080/v1/health

While the session is open, you can point a Tabby IDE extension on your computer at http://localhost:8080. Software Sushi has not tested IDE extensions with this image; see the Tabby documentation for their setup.

Next steps

Attach a model so completions and chat work, in Configuration.

Last updated October 13, 2026 · Suggest an improvement