Key facts
| API compatibility | OpenAI-compatible chat, embeddings and function calling |
| Use cases | Student services Q&A, course content search, research support, document processing |
| Data controls | Scoped API keys, RBAC, SSO, audit logs and region selection |
| Deployment | Plugsky cloud, your VPC, on-prem and air-gapped options |
| Multilingual | 30+ models including multilingual options for campus services |
| Pricing model | Flat monthly self-serve plans with unlimited fair-use usage |
| Free tier | plugsky-micro and plugsky-lite on the free plan, no card required |
| Product status | Chat, streaming, JSON mode, function calling, embeddings, RAG and agents are live |
TL;DR
- Route routine student services questions to grounded Q&A with citations.
- Let students search course materials rather than asking a generic model.
- Keep student records out of prompts; integrate by reference.
- Use on-prem or sovereign deployment where institutional policy requires it.
- Pilot free with one department before a campus-wide rollout.
How it works, step by step
- Pick a narrow service, such as admissions FAQ or library policy Q&A, using published documents only.
- Index the official handbook, calendar and policy pages and require citations in every answer.
- Keep student records and identifiers out of prompts; connect to systems by reference instead.
- Configure region, retention, RBAC and audit logging to match institutional policy.
- Run a documented evaluation with a labelled question set and a human escalation route.
- Move to your own infrastructure with on-prem or air-gapped deployment as policy requires.
Try it yourself
Open the OpenAI-compatible API tester →
University use cases that work today
Universities have large public-document corpora, which makes grounded Q&A the safest and most useful start:
- Student services Q&A: answer questions about deadlines, fees and processes with a link to the official page.
- Course content search: let students query lecture notes and readings they are enrolled in.
- Research support: help with literature triage, extraction and referencing checks.
- Administrative processing: extract fields from forms and transcripts for staff review.
Keep academic judgement and formal determinations with staff.
Student data, research data and policy
Student records, research data and unpublished work fall under institutional and legal controls that vary by jurisdiction. Keep identifiers out of prompts, prefer references to records held in your systems, and set retention to the minimum that supports the workflow. Configure scoped API keys per service, RBAC and SSO for staff, and audit logs so requests are attributable. Choose the processing region deliberately, and move to a VPC, on-prem or air-gapped deployment when policy or a research agreement requires it. The institution remains the data controller; Plugsky provides platform controls.
Serving students without surprising costs
Campus demand is bursty: registration weeks, exam periods and orientation drive sharp peaks. Self-serve plans are flat monthly with unlimited fair-use usage, so peaks do not translate into variable bills, and routing keeps unit costs sensible: a small model for intent classification and short replies, a stronger model for complex multi-document answers. Because the API is OpenAI-compatible, one integration serves the student portal, the library and departmental tools, with scoped keys separating their data. Use plugsky-embed for retrieval so answers stay grounded in approved content.
Rolling out responsibly
Prototype on the free plan with plugsky-micro or plugsky-lite for one department and a public corpus. Use the 14-day full-access trial to benchmark models on real questions, scoring answer accuracy, citation support and escalation rate. Publish an acceptable-use note for students and staff, add a visible escalation path to a human, and measure both resolution and complaint rates. Expand department by department, keeping prompts versioned and re-testing on each model or document change. Where the institution prefers to run its own infrastructure, on-prem deployment keeps the same code.
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 | Often limited regions | You control fully |
| Pricing | Flat monthly self-serve, fair-use usage | Per-token, spikes at term start | GPU plus operations cost |
| Model choice | 30+ models behind one API | Varies by provider | You host every model |
Frequently asked questions
Can we keep our existing OpenAI SDK code?
Yes. Plugsky is OpenAI-compatible, so a pilot built with the OpenAI SDK moves into campus systems with a base URL and model-name change.
Is there a free plan?
Yes. The free plan includes plugsky-micro and plugsky-lite with no credit card, which suits a single-department pilot.
Can students' records be sent to the model?
No. Keep identifiers and records out of prompts and integrate by reference, unless a documented institutional and legal process requires otherwise.
Can we host it on campus?
Yes. On-prem and air-gapped deployment is available, and because the API is compatible, the same application code runs there.
How does pricing work?
Self-serve plans are flat monthly with unlimited fair-use usage, which helps at registration and exam peaks. See the live pricing page for details.
Which endpoints are live today?
Chat, streaming, JSON mode, function calling, embeddings, RAG and agents are live. Audio, images, moderation, files, batch, fine-tuning, assistants and responses are coming soon.
Does the AI grade or admit students?
No. Restrict it to retrieval, drafting and extraction, with academic and admissions decisions owned by staff following institutional process.