Buyer Guide · Updated October 2026

LLM Gateway Pricing: What You Actually Pay

The cheapest advertised gateway is not necessarily the cheapest production setup. Model charges, gateway subscriptions, usage-based fees and engineering operations all matter. Here's how to compare total cost without double-counting.

Practical engineering guidance. Verify provider-specific capabilities, pricing, and support before deployment.

Key takeaways

  • Calculate provider inference and platform costs separately.
  • Check whether gateway fees apply to BYOK traffic, credits, requests or logs.
  • Treat cost-savings estimates as workload-specific; retain quality and success-rate measurements.
  • AetherGate's published Team plan is $249.99/month, Growth $999/month and Scale $2,499/month, with upstream inference billed separately.

What an LLM gateway invoice can include

Most gateways use one or more pricing components: a fixed subscription, percentage of model spend, a fee per request, prepaid inference credits, metered logs or infrastructure costs. Some self-hosted gateways require staff time, compute, storage and ongoing operational support rather than a recurring SaaS subscription.

Bring-your-own-key (BYOK) means your organization has an upstream provider account. It does not guarantee that gateway access is free or that the provider's invoices disappear.

  • Provider invoice: input tokens, output tokens, cache reads and provider-specific tools.
  • Gateway invoice: subscription, overages, seats, requests, logs or usage fees where applicable.
  • Operating expense: hosting, monitoring, configuration and incident response.
  • Migration expense: integration tests, model-name mapping, provider credential rotation and rollback.

A cost comparison formula that prevents double-counting

Monthly total = provider inference + gateway platform fees + gateway-specific metering + operating costs. If a gateway resells inference, do not count the same provider tokens twice. Always calculate at your own volumes and model mix.

For example, a team whose provider inference totals $2,000/month and that chooses a $249.99/month flat subscription has a pre-operations total of $2,249.99/month, assuming no additional platform charges. The example does not demonstrate savings; any savings must be measured against an equivalent direct-provider workload.

What to ask before signing

Request pricing in writing for the workloads you actually run. Verify request rate ceilings, provider connection limits, included retention and how the agreement handles rate-limit failures, provider outages and cancellation.

AetherGate currently publishes Team at $249.99/month, Enterprise Growth at $999/month and Enterprise Scale at $2,499/month, plus provider usage. Validate current entitlements and trial terms on the pricing page before purchase.

  • What happens if we exceed request or workspace limits?
  • Does BYOK incur a percentage markup or request fee?
  • Can we export routing rules, logs and model aliases before cancellation?
  • Are security, audit and support commitments contractual or descriptive?

Verify the cost assumptions

Compare the numbers against OpenAI's published API pricing and AetherGate's current subscription tiers. Then review our gateway architecture guide to include request control and failure overhead.

Sources

Evaluate the gateway on your own workload

Compare an actual gateway request with your direct-provider baseline. Review pricing, security controls and measured behavior before migrating production traffic.

Gateway evaluation resources

Continue from the concept to a practical cost, security or migration decision.

Related gateway comparisons

Frequently asked questions

Are LLM gateways free?

Some offer free tiers or self-hosted software, but provider inference, infrastructure, operations and optional managed features can still cost money.

Does an LLM gateway save money automatically?

No. Savings require evidence from your own workloads after accounting for model quality, retries, latency, provider charges and gateway fees.

Put the advice into practice

Optimize AI work with Foreman

Give the team a job. Foreman weighs the work, model capabilities and price, then explains its choices. Enable web tools for public research. Review, edit or decline proposed actions before they run.

Give Foreman a job →
Technical details

Reasoning-model task routing, persisted specialist/reviewer workflows, web search, Chromium browsing and approval-gated browser/HTTP writes. Fresh public sessions; no signed-in accounts or arbitrary code execution.

Set a limit before the work starts

Per-key spending controls. Each request reserves its estimated cost before it's sent, and parallel calls are only admitted while those reservations fit under the cap.

Set spending controls →
Technical details

Atomic catalog-priced cost reservations, per-key caps, and retained reservations for uncertain charges. Provider invoices, taxes and special rates can differ.