Government Privacy Features Pricing Models Docs Join Waitlist
Now accepting early access signups

Privacy you can prove.
Models you can trust.

The API gateway that lets you control exactly who sees your data. Route to any model. Set your privacy level per request.

Join the Waitlist See How It Works
quickstart.py
from openai import OpenAI client = OpenAI( base_url="https://northerninference.ca/v1", api_key="ni_live_..." ) response = client.chat.completions.create( model="anthropic/claude-sonnet-4.5", messages=[{"role": "user", "content": "..."}], extra_body={"privacy_tier": "managed_canadian_cloud"} )

Route to the world's best models

🇨🇦
Government & Public Sector

Built for Canadian data residency at Tier 3.
Ready for government.

Northern Inference infrastructure runs in AWS ca-central-1 (Montreal). Data at rest stays in Canada. We provide transparent chain-of-custody documentation showing exactly which entities handle your data at each privacy tier, so your security team can make informed decisions.

Canadian Data Residency

Inference runs in ca-central-1 (Montreal) and Azure Canada East. Data at rest stays in Canadian data centres. PIPEDA jurisdiction applies.

Transparent Data Custody

Every API request includes a clear record of which entities processed your data. You choose your privacy tier per request, from customer-hosted hardware (only you see data) to cloud providers with Canadian data residency.

Privacy Tier Model

Select your privacy level per request. Customer-hosted hardware for sensitive workloads, Canadian cloud for general use, or provider-default routes when Canadian data residency is not required.

PII Substitution

Opt-in automatic PII detection and substitution. Names, emails, phone numbers, and other identifiers are replaced with realistic fakes before leaving your infrastructure, then restored in responses. Full audit log available.

Built for Federal Procurement

Inherits AWS Canada's Cloud Service Provider attestations (SOC 2 Type II, ISO 27001, PCI-DSS). ITSG-33 PBMM control mapping in progress. Full status and roadmap on our Trust page.

Procurement-Friendly Pricing

Transparent per-token billing with no hidden costs. Clear invoicing in CAD. Easy to forecast and budget for departmental spending.

Standard API Format

Drop-in replacement for existing AI tools and prototypes. Government developers can adopt without rewriting existing integrations.

Request a Government Briefing

Priority access and dedicated onboarding for government and public sector organizations.

We'll respond within 2 business days. Privacy Policy

PIPEDA Compliant AWS Canada Attestations ca-central-1 by Default Transparent Custody
your_app.py
# Before: locked to one provider from openai import OpenAI client = OpenAI() # single provider # After: one line change, every model client = OpenAI( base_url="https://northerninference.ca/v1" ) # Canadian data residency (Bedrock/Azure) r1 = client.chat.completions.create( model="anthropic/claude-sonnet-4.5", messages=messages, extra_body={"privacy_tier": "managed_canadian_cloud"} ) # Broader model access -> provider-default route r2 = client.chat.completions.create( model="openai/gpt-4.1", messages=messages, extra_body={"privacy_tier": "provider_api"} )

Change one URL.
Add one parameter.

Northern Inference uses the industry-standard chat completions API format. Point any compatible SDK to our endpoint, add a privacy tier, and you're done. No code changes needed.

  • Works with every major AI SDK: Python, Node, Go, and more
  • Route sensitive data through your own hardware via NI. Unlimited tokens, no third parties
  • Switch models without changing code
  • Transparent per-token pricing for every provider
The Key Differentiator

Choose Your Level of Trust

Set your privacy tier on every API call. From zero-knowledge customer-hosted inference to provider-default global access, you control the trade-off.

1
Coming Soon
Customer-Hosted

Self-Hosted Hardware

Your hardware, your premises, open-source models. NI tunnels API requests into your machine; only you ever see prompts and completions. The strongest privacy guarantee we offer. Phase 5 of our roadmap.

Only you process data
2
Coming Soon
Confidential Compute

Nitro Enclave

AWS Nitro Enclaves with cryptographic isolation. Data encrypted with KMS before entering the enclave; AWS itself cannot read inside. Currently CPU-only (small models); GPU support pending Blackwell GA. Phase 5 of our roadmap.

Cryptographic isolation
3
Live
Canadian Cloud

Managed Cloud

AWS Bedrock and Azure OpenAI with Canadian data residency. Providers cannot train on your data. Infrastructure hosted by US-headquartered companies in Canadian data centres (ca-central-1, Canada East) under contractual data protection.

Cloud provider processes in Canada
4
Live
Provider-Default

Global Access

Broader model access through NI-routed upstream providers outside the Canadian-residency boundary. This can include direct APIs, Bedrock in US regions, Azure GlobalStandard, Vertex US regions, and other provider-default routes.

Exact route shown per request
Your Device
→
NI Gateway
→
Canadian Cloud
→
Model

Tier 3: Canadian cloud. AWS Bedrock or Azure OpenAI in ca-central-1. Provider cannot train on your data.

Live on the Platform

Available Models

All models available through the NI API with transparent per-token pricing. Choose by provider, tier, or data residency.

Pre-order Hardware

Register your interest. We'll contact you when your bundle is ready to ship.

Built for Developers

One API. Every Model.

One API for every model. Switch providers with a single parameter.

Standard API Format

Uses the industry-standard chat completions format. Change one URL and your existing code works with 100+ models across providers.

Every Major Provider

Anthropic, OpenAI, Google, Meta, Mistral, Cohere. Access the best model for each task through one gateway.

Transparent Billing

See the provider cost and our fee separately on every request. Thinking tokens billed at the rate shown per model. No expiring credits. No surprise overages.

Per-Request Privacy

Set your privacy tier on every API call. Route sensitive prompts through your own hardware, casual queries through cloud providers.

PII Substitution

Names, emails, and identifiers are replaced with realistic fakes before reaching the model, then restored in responses. Opt-in per API key or per request.

Real-Time Analytics

Live dashboard with per-model, per-team cost breakdowns. Budget alerts and hard limits built in.

Canadian Sovereignty

Infrastructure in Canada (ca-central-1). PIPEDA jurisdiction. Transparent data custody for organizations that need it.

How We Compare

Built Different

Northern Inference is developer infrastructure, not a consumer chatbot.

Northern Inference Venice.ai OpenRouter Direct APIs
Per-request privacy control
Canadian data residency (per request, Tier 3)
Transparent pricing
Standard API format
Per-request custody trail
PII substitution
No crypto required
Simple Pricing

No Surprises. No Games.

Pay for what you use.

See the provider's cost and our routing fee separately on every request. No monthly minimums, no expiring credits, and thinking tokens are priced openly per model.

Transparent per-token rates No monthly minimums Credits never expire Thinking tokens priced openly

Early access members get priority onboarding + founding member pricing

See live per-token rates for every model →

Get Early Access

Be among the first to use Northern Inference.

Early access: priority onboarding + founding member pricing
Start free with CA$5 in credits. No credit card required.

No spam. Unsubscribe anytime. Privacy Policy

Share to move up the waitlist

Each referral moves you closer to the front of the line.