Chroma hosting
A lightweight embedding store that gets out of the way.
Chroma hosting at AIHostBox means a dedicated, single-tenant Chroma 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 $5.99 per month and the admin account, the data and any API keys are yours.
Chroma is an open-source embedding database maintained by the chroma-core team under the Apache-2.0 licence. You create collections, add documents with metadata, and query by vector similarity combined with metadata, full-text and regex filters, from Python, JavaScript/TypeScript or Rust clients. Version 1.0 rewrote the core in Rust, which made it markedly faster without changing the small API that made it popular. It remains the quickest route from notebook to working retrieval, and many RAG tutorials and frameworks use it as their default store.
Out of the box Chroma runs embedded in your Python process, which is fine until a second service needs the same data. Running it as a server — chroma run on port 8000 — gives every application one shared, persistent store, and the index stops living on whichever laptop built it. Chroma keeps metadata in SQLite with the vector index segments beside it in one data directory, so a server with nightly backups is the difference between a demo and something you can restore. Embeddings are still computed client-side or supplied by you.
Chroma 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 Chroma
A single Chroma server container (chromadb/chroma) running chroma run on port 8000, with its data directory — SQLite metadata plus index segments — on persistent storage behind our TLS proxy.
What we configure for you
- Persistent data mounted at /data, the 1.x image default
- A config.yaml for server settings, which 1.x reads instead of the old CHROMA_SERVER_* environment variables
- Token check at the proxy (Authorization or X-Chroma-Token header), following upstream’s proxy-based pattern
- CORS origins in config.yaml if a browser app calls Chroma directly
- chroma utils vacuum after upgrades from pre-1.0 data
Which plan fits
Starter (2 GB) covers prototypes and corpora of a few hundred thousand 384-dimension embeddings. Plus (4 GB) and Pro (8 GB) give room for millions of vectors or higher-dimension models such as 1,536-dimension OpenAI embeddings, which use four times the memory of 384-dimension ones.
The honest limits
Chroma runs here as a single-node server: no replication, no sharding and no user accounts of its own. For very large or multi-tenant workloads, a distributed store is the better fit.
What people run Chroma for
Prototype retrieval
Get a RAG pipeline working before choosing infrastructure.
Small corpora
Tens of thousands of documents without operational overhead.
Shared store for several apps
Let a Flowise flow and a Python service read the same collections.
Filtered retrieval
Narrow similarity search with metadata and document-text filters in the same query.
Licence and version policy
| Item | Position |
|---|---|
| Upstream licence | Apache-2.0 |
| 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.
Reading on Chroma
Works well with Chroma
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 →More Vector & data we host
Qdrant
A fast vector database with real filtering, written in Rust.
from $5.99/moDetails →
Weaviate
A vector database with hybrid search and built-in vectorization modules.
from $11.99/moDetails →
Milvus
Vector search built for very large collections.
from $18.99/moDetails →
Postgres + pgvector
Vector search inside the database you already trust.
from $5.99/moDetails →Chroma hosting: your questions
Does the Chroma server compute embeddings?
How is a hosted Chroma protected when 1.x has no built-in auth?
What changed for existing data when Chroma moved to 1.x?
Can I bring a local persistent Chroma directory?
How big can a collection get on these plans?
Run Chroma without running a server
A dedicated Chroma instance, deployed and maintained by us. Month to month, cancel whenever.