Key facts
| RAG endpoints | POST /v1/embeddings, POST /v1/rag/collections and POST /v1/rag/query |
| Embedding models | plugsky-embed-v1 (1536 dimensions, OpenAI-ada compatible) and plugsky-embed-large (3072 dimensions) |
| Retrieval modes | Keyword, vector and hybrid search with optional cross-encoder reranking |
| Citations | Every query returns ranked chunks with source attribution |
| Ingestion | Documents are chunked, embedded and indexed automatically per collection |
| Batch limits | Up to 2,048 inputs per embeddings request, max 8,191 tokens each |
| Data handling | Per-collection encryption at rest; API data is not used to train models |
| Deployment | Hosted, VPC, on-prem and air-gapped options; 30+ models behind one API |
TL;DR
- Citations make every answer traceable to an authoritative source.
- Air-gapped deployment removes internet egress for classified programmes.
- BYOK and SIEM export support records and audit obligations.
- Pilot on unclassified corpora, then move to the required data plane.
- Keep human review for citizen-facing output.
How it works, step by step
- Classify corpora by sensitivity and decide which deployment plane each one allows.
- Choose a first pilot corpus with high re-use and low classification risk.
- Create collections and ingest documents; verify chunking preserves structure and references.
- Define acceptance in retrieval terms: correct source retrieved and cited on every answer.
- Configure BYOK, audit fields, SIEM export and retention with your security team.
- Review output with subject-matter experts before any citizen-facing use.
- Move to air-gapped or on-prem deployment as classification requires.
Original data
Try it yourself
Open the local RAG stack generator →
Why government need RAG grounded in their own content
Public-sector knowledge work is document-heavy and accountability-heavy: legislation, circulars, procedures, case files and correspondence. Staff need fast answers, but the answer must be traceable to an authoritative source and the processing must satisfy records, privacy and classification rules. A citation-first RAG service fits both requirements: it retrieves from approved corpora and shows where every statement came from.
Start by separating corpora that can be hosted from those that cannot. An air-gapped deployment removes internet egress entirely, with a local model registry and offline update channels for the most sensitive programmes.
The Plugsky RAG stack: embeddings, collections, queries
You do not have to assemble a vector database, an embedder, a retriever and a reranker separately. Documents are chunked, embedded and indexed automatically per collection, retrieval supports keyword, vector and hybrid modes, and every query returns ranked chunks with citations. The endpoints stay OpenAI-compatible — POST /v1/embeddings, POST /v1/rag/collections, POST /v1/rag/query — and the embeddings call also works standalone if you keep your own vector store. For government programmes the endpoint contract stays constant across cloud, VPC, on-prem and air-gapped deployments, which keeps application code stable when the data plane changes.
The controls government should configure first
Sovereignty is delivered through deployment, not promises. Plugsky supports region-locked data planes, private VPC endpoints, on-prem and fully air-gapped installations, so the processing location matches the classification of each corpus. Keys stay under your control with BYOK through AWS KMS, Azure Key Vault, HashiCorp Vault or an on-prem HSM.
Audit logs record model, tokens, latency, user and region per request and export to your SIEM with retention up to 7 years, which supports records-management policy and freedom-of-information workflows. Compliance evidence includes SOC 2 Type II under NDA, ISO 27001/27017/27018, HIPAA with a BAA and GDPR/EU SCC coverage; FedRAMP Moderate is in process, so treat it as pending for US federal work.
A practical pilot path
Run the first pilot on an unclassified corpus with high re-use value, such as internal procedure manuals, and define acceptance in retrieval terms: correct source in the top results and a citation on every answer. Keep humans in the loop for anything citizen-facing. Prototype on the free plan, then move to the required deployment plane before real data is loaded. See the live pricing page for current plans.
Honest comparison
| Capability | Plugsky | Assembling your own stack | Keyword search only |
|---|---|---|---|
| Retrieval | Keyword, vector and hybrid with optional reranking | You integrate and tune each component | Keyword matching, no semantics |
| Citations | Ranked chunks with source attribution | You build attribution yourself | Document links, not passages |
| Embeddings | plugsky-embed-v1 (1536d) and plugsky-embed-large (3072d) | Model choice spread across vendors | Not applicable |
| Deployment | Hosted, VPC, on-prem and air-gapped | Depends on every component | Usually on-prem already |
| Data handling | Per-collection encryption; API data not used to train models | You audit every vendor | Existing controls apply |
| Time to first answer | One collection and a query call | Weeks of integration work | Days, with lower answer quality |
Frequently asked questions
Can RAG run air-gapped?
Yes. Air-gapped deployments run with no internet egress, a local model registry and offline update channels, which suits defence, government and critical-infrastructure programmes.
How do we keep answers accountable?
Every query returns ranked chunks with source attribution, and per-request audit logs tie each answer to a user, model, time and region for later review.
Do we have to change our application code?
No. The RAG endpoints are OpenAI-compatible, so existing SDKs, prompts and evaluation harnesses keep working when you point the base URL at Plugsky.
Is there a free plan?
Yes — plugsky-micro and plugsky-lite with 2 API keys and no credit card. A 14-day full-access trial is also available.
How is pricing structured?
Self-serve plans are flat monthly with unlimited fair-use usage and no per-token charges or overage fees. See the live pricing page for current plans.
Does Plugsky train on our data?
Plugsky states that API data is not used to train models, and collections are encrypted at rest; on Enterprise you can keep everything inside your own VPC.