Onyx hosting
Search and chat across every tool your company already uses.
Onyx hosting at AIHostBox means a dedicated, single-tenant Onyx instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $27.99 per month and the admin account, the data and any API keys are yours.
Onyx, which started life as Danswer, is an open-source AI platform that pairs a chat assistant with enterprise search over your own tools. Its connectors index wikis, ticketing systems, shared drives and chat history from more than 50 applications into a hybrid keyword and vector index, and answers cite the passages they came from. Agents and actions, including MCP tools, sit on top, and it works with the major hosted LLM providers or with self-hosted models.
That breadth is why it is a server application. The standard deployment runs around ten containers: the API server and background workers, a web front-end, two model servers (one for indexing, one for queries), Postgres, OpenSearch, Redis, MinIO and nginx. The background workers re-sync connectors on a schedule, which a laptop that sleeps cannot do. The Community Edition is MIT-licensed; SSO, RBAC, usage analytics and query history belong to the separately licensed Enterprise Edition.
Onyx hosting plans
Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.
How we run Onyx
Onyx’s official Docker Compose stack: api_server and background workers from the onyx-backend image, the web_server, separate inference and indexing model servers, Postgres, OpenSearch, Redis, MinIO for file storage and nginx in front.
What we configure for you
- AUTH_TYPE=basic with your first admin account created
- SMTP settings so invitations and password resets are delivered
- Your LLM provider key, or an Ollama or OpenAI-compatible endpoint, in the admin panel
- FILE_STORE_BACKEND=s3 pointed at the bundled MinIO
- Connector credentials (Confluence, Google Drive, Slack, Jira and others) added with you on the first sync
- Embedding model choice, including a hosted embedding provider to lighten the model servers
Which plan fits
Onyx documents 10 GB of RAM as the minimum for its standard, connector-indexing mode, so both plans start above it. Pro (12 GB, 6 vCPU) suits a team with a handful of connectors; Max (16 GB, 8 vCPU) gives OpenSearch and the two model servers room for several large sources syncing at once.
The honest limits
SSO, RBAC and analytics are Enterprise Edition only. The first full index of a large wiki or drive can take hours on CPU, because embeddings are computed on the same cores that serve queries unless you use a hosted embedding model.
What people run Onyx for
Internal answer engine
Stop asking colleagues where a document lives.
Support deflection
Answer repeat questions from the existing knowledge base.
Slack answer bot
Let colleagues ask in a channel and get cited answers without opening another tool.
Agents with company context
Build assistants that search indexed sources and call actions or MCP tools in one conversation.
Licence and version policy
| Item | Position |
|---|---|
| Upstream licence | MIT (core) |
| Sold as managed hosting | Yes |
| Tenancy | Single tenant — one customer per instance |
| Patching | Critical advisories within 24 hours, high severity within 72 |
Our full position on licences, including the applications we decline to sell, is on the licensing page.
Works well with Onyx
Ollama
Your private LLM server — pull a model, get an API, keep every token on your own box.
from $16.99/moDetails →
LiteLLM
One OpenAI-compatible endpoint in front of every model provider you use.
from $5.99/moDetails →
MCP Gateway
Host your Model Context Protocol servers where your agents can always reach them.
from $5.99/moDetails →More Agents & RAG we host
Dify
A visual studio for building LLM apps, agents and RAG pipelines.
from $18.99/moDetails →
Flowise
Drag-and-drop LLM flows, agents and chatbots you can embed anywhere.
from $11.99/moDetails →
Langflow
A visual builder for LangChain flows and multi-agent systems.
from $11.99/moDetails →
AnythingLLM
A private document workspace — drop files in, ask questions, get cited answers.
from $18.99/moDetails →Onyx hosting: your questions
Which LLMs can Onyx use?
Does the Community Edition include SSO?
How much disk does indexing use?
What is backed up in an Onyx instance?
Can I move an existing self-hosted Onyx or Danswer install here?
Run Onyx without running a server
A dedicated Onyx instance, deployed and maintained by us. Month to month, cancel whenever.