Requesty
Sign in

Pricing

How much does Requesty cost?

Free on the free models, with 200 requests a day and no card. After that, 5% on top of what the model costs. No seats, no tiers, no minimum.

Free

$0

no card, no trial clock

The whole platform, on the free models.

  • Every free model in the catalogue
  • 200 requests a day
  • Routing, caching and fallbacks
  • Spend tracking and analytics
  • EU data residency

Most teams start here

Pay as you go

5%

on top of model cost

Everything, priced off what your application spends.

  • All 600+ models
  • Bring your own provider keys
  • Routing policies, caching and failover
  • Spend limits and budget caps
  • MCP gateway
  • Advanced observability
  • EU data residency
  • Email support

No seat fees, no feature tiers, no minimum spend.

Enterprise

Volume

discounts and committed spend

For the teams whose security review comes before their first call.

  • SSO: Okta, Azure AD, Google Workspace, custom OIDC
  • Full RBAC and audit logs
  • Approved model allowlists
  • Teams, groups and team budgets
  • Guardrails and PII detection
  • Service accounts for CI/CD
  • Dedicated support and custom SLAs

EU data residency on every plan  ·  Email sales@requesty.ai

01Compare

Every line, all three plans

The whole matrix, including the rows where the answer is no and the one where it is not yet.

Features by plan: Free, Pay as you go, and Enterprise
FeatureFreePay as you goEnterprise
Models and providers
Free modelsYesYesYes
All 600+ modelsNoYesYes
200 requests a dayYesNo limitNo limit
20+ providersYesYesYes
Bring your own keysNoYesYes
Approved model allowlistNoNoYes
Custom model hostingNoNoYes
Routing and reliability
Routing policiesYesYesYes
Automatic fallbacksYesYesYes
Load balancingYesYesYes
Latency routingYesYesYes
Prompt cachingYesYesYes
Custom routing rulesYesYesYes
Security and compliance
EU data residencyYesYesYes
GuardrailsNoNoYes
PII detection and scrubbingNoNoYes
SSO: Okta, Azure AD, Google Workspace, custom OIDCNoNoYes
Full RBACNoNoYes
Audit logsNoNoYes
SOC 2 Type IIObservation period under way with an independent auditor. Not a plan feature, and not yet a certificate.In progressIn progressIn progress
Teams and access
Spend limits per keyYesYesYes
Service accounts for CI/CDYesYesYes
User managementYesYesYes
Groups and team budgetsNoNoYes
Key management APINoNoYes
Analytics and monitoring
Real-time analyticsYesYesYes
Cost trackingYesYesYes
Tool call analyticsYesYesYes
Session reconstructionYesYesYes
Spending alerts, Slack and webhooksYesYesYes
Custom dashboardsYesYesYes
Data exportsYesYesYes
Advanced observabilityNoYesYes
Developer tooling
MCP gatewayYesYesYes
Prompt libraryYesYesYes
Prompt optimisationYesYesYes
Structured outputsYesYesYes
Reasoning and extended thinkingYesYesYes
Support
Community supportYesYesYes
Email supportNoYesYes
Dedicated supportNoNoYes
Custom SLAsNoNoYes

Looking for per-model prices? The model catalogue lists every rate, and the cost calculator works out a month from your own token volumes.

02What is not on the bill

The line items that are not there

On most gateways the price is the markup plus four other things. Here it is the markup.

Per seat
Nothing per user

Add your whole team, and the ones who only read the dashboard. The bill follows tokens, not headcount.

Per feature
No feature tiers

Routing, caching, failover, analytics and audit logs are on every paid account. There is no plan that unlocks them.

Per month
No minimum

No commitment and no platform fee. A month where you send nothing costs nothing.

Per region
EU residency included

Frankfurt, on AWS eu-central-1, at the same 5%. Residency is not an upgrade.

03Your own keys

Already have provider contracts? Bring them

Keep the rates you negotiated and still route, cap and observe every call through one endpoint.

Bring your own keys
0%
on traffic through your own contracts

Your keys, your negotiated pricing, your invoices from the provider. Requesty still applies routing policies, spend limits, caching, guardrails and audit logs across all of them, so the governance does not have to be rebuilt per provider.

How BYOK works
MixedBoth at once
your keys where you have them

Use your own contract for the two providers you have negotiated and the Requesty catalogue for the other eighteen. One key to your application either way.

VolumeEnterprise
discounts and committed spend

Committed spend, annual billing, private regions, a signed DPA and a named contact. Procurement handled by people rather than a form.

Speak to a founder

Questions about the bill

Is there a free tier?

Yes. The free tier is the whole platform on the free models: routing, caching, fallbacks, analytics and EU data residency, with a limit of 200 requests a day and no card. Move to pay as you go whenever you want the 600+ paid catalogue and your own provider keys.

How does the 5% markup work?

We add 5% to the provider cost. A model that costs $10.00 per 1M tokens costs $10.50 through Requesty. There is no other fee.

Can I bring my own API keys?

Yes. Use your own provider keys and keep the pricing you have negotiated, while still getting routing policies, spend limits, observability and guardrails across all of them.

Where is my data stored?

EU data residency is available on every plan, at no extra cost. Traffic, logs and analytics stay on EU infrastructure in Frankfurt, on AWS eu-central-1, via router.eu.requesty.ai.

Is there a minimum commitment?

No. Pay as you go has no minimum and no platform fee. You pay for what you use, and a month with no traffic costs nothing.

Do you offer annual billing?

Enterprise agreements can be structured with annual billing and committed spend. That is a conversation rather than a checkout page.

What payment methods do you accept?

All major credit cards. Enterprise customers can be invoiced by ACH or wire transfer.

Model catalogue  ·  Docs  ·  EU residency  ·  Security

Start on the free tier

200 requests a day, no card. Add a payment method when you outgrow it.

Or email sales@requesty.ai.