Hybrid in a container environment you run

Services you run in containers, on your own server or in your own cloud account

Words in the cells: YesA page we read says so., PartlyA page says part of it, for example separate models per agent without saying the providers can differ., Via a proxyOnly with a proxy, or a local server that imitates a cloud provider's interface, in front of the tool., NoA page we read says the product does not support it., ClaimedA vendor page says so on a marketing or overview page, and we found no guide showing how to set it up., Not foundWe read pages for it and found nothing there. It says nothing about whether the product has the feature., n/aThe question does not apply to the product, for example because it is a model server and not a tool that connects to models., BlankNothing is published for it yet.

A container environment you run can sit in several places: your own laptop running Docker, a server in your office, or a virtual machine or a Kubernetes cluster in a cloud account you control. In each case you run the software and choose where it sits, and no vendor runs it for you. The figure shows the step a service takes before it answers.

The container route each one names

• LibreChat: its pages name Docker and Helm (www.librechat.ai page, read 2026-10-22)
• LiteLLM: its pages name Docker (docs.litellm.ai page, read 2026-10-22)
• Open WebUI: its pages name Docker, Kubernetes and Helm (github.com page, read 2026-10-22)
• OpenHands: no container route is named on the pages we read
• Stack AI: no container route is named on the pages we read

Model servers that you host are in the table of inference servers further down.

Products that run in more than one place

These come in forms that run in more than one place, so each is on every page that fits one of its forms.

Where their models run

Where each product's models run, as its vendor's pages state it.

Inference servers you host

An inference server loads a model and answers requests from the tools on the other pages. You run it yourself, in a container, on a GPU server or in a cloud account you control. The table shows what each server's pages say about the two formats a tool can speak to it, and whether its pages cover Claude Code.

Servers that run in more than one place

• llama.cpp server: on this page as model server. Also as command-line tool, on the page Hybrid on your own machine.
• Ollama: on this page as model server. Also as desktop app and command-line tool, on the page Hybrid on your own machine.

Model servers we have not read for these questions: Atomic-Chat, BentoML, foundry-local, model-runner, sglang.