Pricing

Transparent pricing
for production AI

Every plan includes intelligent routing, observability and built-in security controls — Shield, DLP and the full output-safety suite — from day one. A seat is a dashboard or API user, not every employee. Scale with seats and routed usage as your platform grows. No hidden platform fees, no surprise feature gates.

Per-seat billing, monthly Cancel anytime Beta customers lock in their price

Starter

$29 / seat / month

For teams exploring AI routing and reducing LLM spend.

5 seats 5,000 routed requests / seat $0.50 per additional 1,000 requests

Smart provider & model routing
Automatic failover & retries
Cost & latency analytics
Prompt injection protection
DLP & PII masking

Start free trial

Recommended

Pro

$49 / seat / month

For teams running production AI workloads that need visibility, governance and cost control.

25 seats 10,000 routed requests / seat $0.25 per additional 1,000 requests

Includes everything in Starter
Advanced routing conditions
Team & project isolation
Spend budgets & alerts
SAML SSO
Priority email support

Start free trial →

Enterprise

Custom

For organisations operating AI across multiple teams, regions and regulated environments.

Custom seat commitments Pooled routed usage Volume-based pricing

Includes everything in Pro
99.9% SLA-backed uptime
SLA & compliance options
SCIM provisioning
Extended audit log retention
SIEM & compliance integrations
Dedicated onboarding & support

Book a call →

A seat is a dashboard user — an admin, developer or analyst who signs in to configure routing, policies and budgets. Employees using routeur.ai Chat are licensed separately under the Chat platform licence and do not consume gateway seats.

Feature	Starter $29 / seat / mo	Pro $49 / seat / mo Recommended	Enterprise Custom
Plans & seats
Seats included	5	25	Custom commitments
Per-seat price	$29 / mo	$49 / mo	Negotiated
Annual discount	—	10% off	Custom
Routing & gateway
Routed requests included	5,000 / seat / mo	10,000 / seat / mo	Pooled, negotiated
Smart provider & model routing	✓	✓ Advanced conditions	✓ Advanced conditions
LLM providers	All providers	All providers	All + custom integrations
Automatic failover & retries	✓	✓ Multi-provider	✓ Multi-region
Team & project isolation	—	✓	✓
Overage — routed requests	$0.50 / 1K	$0.25 / 1K	From $0.10 / 1K
Inbound safety applied to prompts
Shield Prompt-injection & abuse defence	10,000 / mo then $0.40 / 1K	100,000 / mo then $0.30 / 1K	Pooled Negotiated
DLP PII & secrets redaction	10,000 / mo then $0.80 / 1K	100,000 / mo then $0.60 / 1K	Pooled Negotiated
Outbound safety applied to responses
Toxicity Hateful or harassing language	5,000 / mo then $0.80 / 1K	50,000 / mo then $0.60 / 1K	Pooled Negotiated
Output Moderation Response-side prompt shielding	5,000 / mo then $0.80 / 1K	50,000 / mo then $0.60 / 1K	Pooled Negotiated
Observability & cost control
Cost & latency analytics	✓	✓	✓
Spend budgets & alerts	—	✓	✓
Per-SKU spend caps	✓	✓	✓
SIEM & compliance integrations	—	—	✓
Security & compliance
SAML SSO	—	✓	✓
SCIM provisioning	—	—	✓
Audit log retention	7 days	90 days	Extended & configurable
Data residency	✓	✓	✓ + region pinning
Support
Channel	Standard email	Priority email	Dedicated onboarding & Slack
Response SLA	Best-effort	Next business day	1 hour
Uptime SLA	—	99.5%	99.9% SLA-backed
Get started	Start free trial	Start free trial →	Book a call →

All add-on SKUs are billed per call, per route. Set spend caps per workspace, per route or per API key — overages can never run away. Prices in USD, exclusive of VAT/tax.

Current AI API spend What you pay providers today, per month

/ month

Share of traffic that's routine Classification, extraction, short summaries, formatting — work a small model handles just as well

% of requests

Seats People who sign in to manage routing, policies, keys and budgets — plus developers using the API. Not every employee.

seats

Net saving, after paying us

$1,080

$12,960 a year

Routing saving: $1,225
Starter · 5 seats: −$145
Net per month: $1,080

$5,000 × 35% routine × 70% cheaper = $1,225 saved a month, against $145 for 5 Starter seats — routing covers the platform 8.4× over.

Start your free trial → Book a call

Routine requests moved from a frontier model to a small one typically cost around 70% less; we apply that reduction only to the share of traffic you set, and to nothing else. Your real figure depends on your workload — the dashboard reports actual savings against going direct, per request, once you're live. Seat pricing is list price; usage above your plan's pooled allowance is billed separately at the rates on this page.

What counts as a "routed request"?

What counts as a seat?

A seat is a member of your routeur.ai workspace: the people who sign in to the dashboard to manage routing rules, policies, credentials and budgets, plus developers using the API. You do not need a seat for every employee in your organisation. Company-wide access to the governed chat workspace at chat.routeur.ai is priced separately as a platform licence plus token usage at cost — see the Chat page or talk to us.

How is Chat priced differently from the gateway?

Two different motions. The gateway (Starter / Pro / Enterprise above) is billed per seat plus routed usage — seats are the admins, developers and analysts who configure routing, policies and budgets. routeur.ai Chat is for your whole workforce, so it's priced as a platform licence plus token usage at cost rather than per gateway seat — every employee gets every approved model behind your policies, DLP and audit trail without a per-person AI subscription. If you have 200 employees and 4 platform engineers, you'd license those 4 engineers as gateway seats and cover the 200 employees under the Chat platform licence. See the Chat page or talk to us for a quote.

How do per-seat allowances work?

Every seat adds its allowance to one shared monthly pool for your organisation — 5,000 routed requests per seat on Starter, 10,000 on Pro. A fully-seated Starter workspace (5 seats) pools 25,000 requests a month; a fully-seated Pro workspace (25 seats) pools 250,000. Usage is metered against the pool, not against individuals, so one busy service account doesn't need its own plan. Once the pool is used, overage pricing applies per 1,000 requests and appears as a separate line item on your next invoice.

Are the safety SKUs really included from day one?

Can I cap spend so a runaway loop doesn't bankrupt me?

Do unused allowances roll over?

Can I switch plan mid-month?

What if I'm in private beta?

Transparent pricing
for production AI

A governed AI workspace for your whole team.

Every limit, every SKU, side by side.

What routing would be worth to you.

Pricing FAQ

Lock in your price today.

Transparent pricingfor production AI

Plans

A governed AI workspace for your whole team.

Every limit, every SKU, side by side.

What routing would be worth to you.

Pricing FAQ

Lock in your price today.

Transparent pricing
for production AI