Key facts
| API compatibility | OpenAI-compatible /v1/chat/completions; drop-in base URL change |
| Models | 30+ models behind one API; open-weight options for private deployment |
| Typical patterns | Citizen Q&A, case summaries, multilingual comms, policy research |
| Integration path | Connects to case-management, CMS and contact-centre systems through middleware |
| Pricing model | Flat monthly self-serve plans with unlimited fair-use usage; no per-token billing — see the live pricing page |
| Free tier | plugsky-micro and plugsky-lite on the free plan, no card required |
| Deployment | Plugsky cloud, your VPC, on-prem and air-gapped options |
| Compliance posture | SOC 2 Type II and ISO 27001 readiness in progress (not yet certified); validate evidence during diligence |
TL;DR
- Use the API to assist staff and inform citizens; humans decide.
- Choose sovereign or region-locked deployment for resident data.
- Meet accessibility and language requirements in generated content.
- Keep records of prompts, outputs and the policy basis used.
- Start free with plugsky-micro and plugsky-lite, no card required.
How it works, step by step
- Select an internal workflow such as policy Q&A or document summarisation.
- Complete privacy, records and accessibility reviews before the pilot.
- Classify data and keep sensitive case material out of early phases.
- Build against the OpenAI-compatible endpoint with scoped keys.
- Configure residency, SSO and audit export per procurement requirements.
- Pilot, measure accuracy and escalate paths, then consider citizen-facing use.
Original data
Try it yourself
Open the AI data residency checklist →
Where an AI API fits in government
Public-sector value comes from helping staff and citizens find and understand information, not from automating decisions. Sensible starting points:
- Citizen Q&A: answer questions from published guidance with citations, and hand complex cases to staff.
- Case-file summaries: condense long case histories into a structured summary for the caseworker who decides.
- Multilingual communication: draft notices and responses in the languages your population speaks, for human review.
- Policy research: summarise legislation, guidance and consultations into briefing notes with references.
- Back-office correspondence: draft routine letters and responses from templates and verified facts.
Security, privacy and data handling
Public trust depends on lawful, explainable and reviewable processing, so governance precedes deployment:
- Keep automated decisions with legal effect out of scope; humans decide, models assist.
- Choose sovereign or region-locked deployment where data must remain in-country.
- Meet accessibility and language requirements for anything citizen-facing.
- Retain audit logs and record the policy basis for each use case.
Deployment options and model choice
Sovereignty requirements usually mean a region-locked plane or an in-country private deployment, with audit export wired into your records. The same OpenAI-compatible API runs across Plugsky cloud, a private endpoint in your VPC, on-prem and air-gapped, with region-locked planes for residency. One key reaches 30+ models, including open-weight options for offline deployment, and migration is a base URL change. Chat, streaming, JSON mode, function calling, embeddings, RAG and agents are live; audio, images, moderation, files, batch, fine-tuning, assistants and the responses API remain coming soon. The free plan includes plugsky-micro and plugsky-lite with no card, and a 14-day full-access trial covers paid tiers — see the live pricing page for current plans.
From pilot to production
Public-sector failures attract scrutiny disproportionate to their size. Avoid:
- Piloting with live case files before a privacy and records review.
- Putting a chatbot in front of citizens before accuracy and escalation paths are proven.
- Ignoring accessibility standards in generated content.
- Assuming a vendor certification satisfies your own statutory obligations.
- Skipping the records-retention analysis for prompts and outputs.
Start internally with policy Q&A and summarisation, involve privacy, records and accessibility teams early, and publish a clear escalation path for anything citizen-facing. Keep the human decision point documented for every workflow.
Honest comparison
| Capability | Plugsky | Typical per-token API | Building in-house |
|---|---|---|---|
| API compatibility | OpenAI-compatible chat, embeddings and tools | Usually compatible | Full rewrite |
| Deployment | Cloud, VPC, on-prem and air-gapped | Mostly cloud-only | You operate GPUs and serving |
| Data residency | Region selection and sovereign options | Limited regions | You control fully |
| Pricing | Flat monthly self-serve, fair-use usage | Per-token, harder to forecast | GPU plus operations cost |
| Model choice | 30+ models behind one API | Varies by provider | You host every model |
| Industry fit | Citizen Q&A, case summaries, multilingual comms, policy research | Generic API, you adapt it | You build every workflow |
Frequently asked questions
Can we keep our existing OpenAI SDK code?
Yes. Plugsky exposes an OpenAI-compatible API, so you change the base URL and model name and keep your integration.
Is there a free plan?
Yes. The free plan includes two free models, plugsky-micro and plugsky-lite, and does not require a credit card.
Can data be kept in-country?
Region-locked planes keep prompts, storage and logs in a selected region, and enterprise options include VPC, on-prem and air-gapped deployment.
How does pricing work?
Self-serve plans are flat monthly with unlimited fair-use usage; enterprise agreements cover residency, capacity and SLA terms. See the live pricing page for current plans.
Is Plugsky certified?
SOC 2 Type II and ISO 27001 readiness are in progress rather than completed. Record that accurately and validate controls during procurement.
Which endpoints are live today?
Chat completions, streaming, JSON mode, function calling, embeddings, RAG and agents are live. Audio, images, moderation, files, batch, fine-tuning, assistants and the responses API are coming soon.
Should citizens interact with AI directly?
Only with strict grounding, clear disclosure, accessibility and a visible escalation path. Many agencies start with staff-facing assistance first.