Self-hosted software, hosted right
This section covers self-hosted alternatives to popular SaaS: the projects, the deployment guides, and the trade-offs. Every guide ships a single docker-compose.yml you can copy-paste, a measured RAM/CPU footprint, an honest "when not to use this" section, and a recommended hosting tier on Liquid Web or Nexcess depending on workload.
Pricing in linked guides is subject to change; the verification date is shown inline on every price.
Categories
- AI / LLM ops: Ollama, vLLM, LiteLLM, Open WebUI, Dify, Flowise, AnythingLLM, LibreChat
- AI assistants: OpenClaw, NemoClaw, Hermes Agent
- CRM & GTM: Twenty CRM, Mautic, Billionmail, Listmonk, Postal, Chatwoot, Cal.com, Formbricks
- Analytics: Plausible, Umami, PostHog
- Identity: Authentik, Keycloak, Pocket ID
- DevOps & PaaS: Coolify, Dokploy
- Productivity: Outline, Plane, Karakeep
- Media & storage: Immich
How we test
Every singleton and combo is deployed locally on Docker with a host sized to match the recommended Liquid Web tier (e.g. 8 vCPU / 8 GB RAM for the 8 GB Managed VPS recommendation). Footprint readings disclose this exactly: equivalent specs, expected to behave the same on Liquid Web.
Hosting recommendations
For pure infrastructure (VPS, dedicated, GPU, HIPAA, HA clusters, managed databases), we recommend Liquid Web. For managed application stacks (WordPress, WooCommerce, Magento, Drupal), we recommend Nexcess, Liquid Web's sister brand with an app-grade focus. See Liquid Web vs Nexcess: which brand is right for you for the full routing matrix.
