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 | Proposal drafts, research synthesis, note summaries, knowledge search |
| Integration path | Connects to knowledge management, CRM and document repositories via 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 |
| Identity and audit | SSO/SCIM, role-based access, SIEM audit log export and BYOK via KMS or HSM |
TL;DR
- Use the API for drafting and synthesis; keep judgement with consultants.
- Isolate client knowledge per engagement with separate keys and indexes.
- Exclude client content from training in the DPA and verify it.
- Verify every generated figure against the source document.
- Start free with plugsky-micro and plugsky-lite, no card required.
How it works, step by step
- Start with internal material such as proposal templates and public research.
- Define engagement-level isolation for keys, indexes and access control.
- Agree the DPA terms for training exclusion, retention and subprocessors.
- Build against the OpenAI-compatible endpoint and version your prompts.
- Pilot on one engagement with client consent and a human review gate.
- Measure drafting hours saved and accuracy, then expand practice by practice.
Try it yourself
Where an AI API fits in consulting
Consulting sells expertise and time, so the wins come from removing drafting and search overhead while protecting client confidentiality:
- Proposal and RFP drafts: assemble a structured first draft from prior proposals, CVs and templates for the partner to rewrite.
- Research synthesis: condense market reports and interview transcripts into themes with source references.
- Interview and workshop summaries: turn raw notes into minutes, decisions and open questions within hours.
- Deliverable formatting: restructure rough analysis into house templates consistently across teams.
- Knowledge search: find prior engagement methods and benchmarks, subject to ethical-wall rules.
Security, privacy and data handling
Client confidentiality and conflicts rules make engagement isolation the central design requirement:
- Separate projects and keys so one client's content cannot surface in another's retrieval.
- Apply access controls that mirror your ethical-wall policy, not just folder permissions.
- Exclude client content from training in the DPA and verify the implementation.
- Keep audit logs of who queried what so confidentiality reviews are answerable.
Deployment options and model choice
Engagement isolation is easier to enforce in a private deployment, but region-locked cloud with per-client keys covers many firms. 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
Confidentiality failures in consulting are reputational events. The usual causes are mundane:
- Indexing all client documents into one shared retrieval space.
- Reusing prompts and keys across engagements with different terms.
- Summarising interview notes that still contain unnecessary identifying detail.
- Letting generated proposals carry figures nobody verified.
- Skipping the DPA because the pilot used internal material.
Pilot on internal knowledge, then one engagement with explicit client consent and isolated retrieval. Verify numbers against source documents, keep an audit view per engagement, and only then standardise the workflow across the firm.
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 | Proposal drafts, research synthesis, note summaries, knowledge search | 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 and prompts.
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.
How does pricing work?
Self-serve plans are flat monthly with unlimited fair-use usage; enterprise agreements cover isolation, residency and SLA terms. See the live pricing page for current plans.
Will client data be used for training?
Plugsky processes customer data to provide the service. Training exclusion belongs in your DPA; confirm current terms with the enterprise team before indexing client material.
Can we keep engagements separated?
Yes. Use separate projects, keys and retrieval indexes per engagement, and mirror your ethical-wall rules in access control.
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.
How do we handle figure-heavy deliverables?
Treat generated numbers as drafts and verify them against the source document. Restrict models to restructuring and synthesis where accuracy is contractual.