Skip to content
Docs menu

Deploy

Run venom's orchestrator and console on your own hardware, from one laptop to a Kubernetes cluster or an air-gapped site, and build and verify releases.

View .md

venom's server side is two programs: the orchestrator (Go, an OpenAI-compatible gateway on port 8787 in front of your model servers and provider keys) and the console (Next.js, the web UI over the orchestrator's admin API). The CLI talks to the orchestrator; the console is optional. Every way of running them below keeps the same rules:

  • Loopback first. Every published port binds 127.0.0.1 unless you change it. Before you expose the orchestrator or the console on another interface, set real values for VENOM_API_KEY, VENOM_ADMIN_TOKEN and the console password, and put TLS in front.
  • Nothing calls out by default. No telemetry and no update checks. The only outbound connections are the ones you configure: provider upstreams with keys, mesh mode (VENOM_CLOUD_URL), and the desktop daemon's auto-update, which is off until you turn it on.
  • Keys stay in the environment. In .env for compose, in Secrets on Kubernetes, in a 0600 file for the desktop daemon. Nothing on these pages prints one.

Which one to use

You have Use Needs
A laptop or desktop, no Docker The desktop daemon (venom-daemon) nothing; the installer brings its own Node.js
A machine with Docker and a venom checkout Docker Compose (make up) Docker, make
A GPU server you want to share Docker Compose (make daemon) Docker, a GPU
A Linux server for Venom Cloud Install Venom Cloud on a Linux server (one command, HTTPS) Docker, or the installer adds it
A Mac, Venom Cloud for yourself Install Venom Cloud on a Mac (http://localhost:8080) Docker Desktop, OrbStack or Colima
An AWS account Venom Cloud on AWS (CloudFormation stack) an AWS account
A Kubernetes cluster Kubernetes (kustomize manifests) kubectl
A rented GPU on RunPod or Lambda Cloud Terraform BYOD GPU Terraform, a cloud API key
No network at all Air-gapped install (make airgap) a connected build machine, once
Your own release pipeline Releases and signatures (make release) Go, Node, Docker

Monitoring for any of them is on Observability; the console's settings are on The web console.

In this section

  • Docker Compose and profiles: Run the orchestrator and console with docker compose from a venom checkout, add local inference, metrics or web search with profiles, and share a GPU box with make daemon.
  • The web console: Configure the venom web console's login, reverse-proxy trust, connection to the orchestrator and venom serve, and the data files behind its Flow and Scorecards pages.
  • Observability with Prometheus and Grafana: Run Prometheus and a provisioned Grafana dashboard next to the orchestrator with the observability compose profile, or scrape the orchestrator's /metrics from your own monitoring.
  • Kubernetes: Run the orchestrator on Kubernetes from venom's kustomize manifests, with keys in Secrets and the config as a generated ConfigMap, optionally as a Venom Cloud daemon.
  • Helm chart for Kubernetes: Install Venom Cloud and the orchestrator on an existing Kubernetes cluster with one Helm chart, your own or a bundled Postgres, generated secrets, a pre-upgrade migration Job, ingress with TLS and optional NetworkPolicies.
  • Self-host Venom Cloud: Run the Venom Cloud workspace on your own server with Docker Compose, your own Postgres and a built-in cron runner, with no Vercel or Neon.
  • Install Venom Cloud on a Linux server: One command that checks your server, generates the secrets, starts Venom Cloud with HTTPS (on one server, or with your own database server and daily backups) and leaves a venom-cloud command for status, logs, backups, restores and upgrades.
  • Deploy Venom Cloud on AWS: Create one CloudFormation stack in your own AWS account. Standard is one encrypted EC2 instance with HTTPS and nightly snapshots; Production is an Auto Scaling group behind a load balancer, RDS Postgres with pgvector, Secrets Manager and an S3 bucket for backups.
  • Install Venom Cloud on a Mac: Run Venom Cloud on your own Mac from the command line, at http://localhost:8080, in Docker Desktop, OrbStack or Colima, with no root, started again at each login by a LaunchAgent.
  • Update a self-hosted Venom Cloud: Check for a signed release, upgrade with a preflight, a database backup, a health check and automatic rollback, roll back from the console or by hand, switch on automatic updates in a maintenance window, use the preview channel, and upgrade an air-gapped server from an offline bundle.
  • Back up a self-hosted Venom Cloud: Scheduled backups of the database and the server's files to its own disk or your S3-compatible bucket, a verify restore into a scratch database, and a full export to download, all from Settings, Server backups.
  • Link a self-hosted install to Venom for managed service: Connect a self-hosted Venom Cloud to Venom's management plane with a one-time code, read exactly what its hourly heartbeat sends, disconnect at any time, and how Venom's operators see linked installs.
  • The desktop daemon: Install venom-daemon on a desktop or GPU box with no Docker, manage it from the command line, the dashboard or the macOS menu bar, connect it to Venom Cloud, and turn on opt-in auto-update.
  • Projects on the desktop daemon: The folders on your computer that coding agents can work in, from the daemon's dashboard or with venom-daemon projects; the workspaces Venom Cloud makes for chats; nothing else is served.
  • Set up "Connect with Meta" for WhatsApp: Everything the operator does in Meta's developer console so Settings, Comms offers Connect with Meta (WhatsApp Embedded Signup and coexistence): the app, Tech Provider, app review, the login configuration, domains, the webhook and four environment variables.
  • Terraform BYOD GPU: Rent a GPU box on RunPod or Lambda Cloud with venom's Terraform module, install the released venom-daemon with Ollama and your models on it, and join it to Venom Cloud.
  • Air-gapped install: Build one verified archive of venom on a connected machine with make airgap, carry it to a machine with no network, install it with checksum and signature checks, and bring your own model.
  • Releases and signatures: Verify a venom download against SHA256SUMS and its minisign signature, and build your own release, library packages, SBOM and other artifacts from a checkout with make release, make libs and related targets.
  • The venom-computer host image: Build and run the host container for Venom Computers, with venom-seatd, the seat endpoints, the environment it reads, how to run it locked down, and what a seat costs in memory.
  • Venom Computers on a daemon: Turn on persistent agent browsers on your own daemon, and see how many seats it holds, how many run at once, what happens beyond that, how idle seats hibernate, and what the daemon tells the cloud.