Pricing & Billing

Clear Pricing Before You Route Production Traffic

Pricing is more than a model table. Compare model prices, understand pay-as-you-go versus BYOK economics, and see exactly how usage translates into spend.

Transparent Pricing

Compare current model pricing for pay-as-you-go and BYOK options.

View all pricing
anthropic
Claude Sonnet 4.6
Input$3.24 / 1M tokens
Output$16.20 / 1M tokens
Secure checkout

Payments are processed through Stripe Checkout.

Available card and wallet options vary by region and eligibility. The final payment methods are shown in checkout before you confirm the purchase.

VISAMastercardGoogle PayStripe

Plans for every Claude workload

Pick the plan that matches how much Claude traffic you route. Monthly or yearly billing, cancel anytime, usage credited to your account.

Starter

$89 / month
For solo builders putting their first Claude traffic through the API.
Cancel anytime.

Includes

  • Claude models behind one OpenAI-compatible endpoint
  • Health-aware routing with automatic failover
  • Per-key usage and spend visibility
  • Bring your own provider keys (BYOK)
  • Email support

Standard

$169 / month
For products in production that need headroom and failover.
Most teams start here.

Everything in Starter, plus

  • Higher monthly usage allowance
  • Automatic failover across approved routes
  • Spending limits and budget alerts
  • Team keys with per-key attribution
  • Priority email support

Premium

$249 / month
For high-volume workloads and multi-team governance.
Volume pricing available.

Everything in Standard, plus

  • Highest monthly usage allowance
  • Cost-aware routing with approved model baskets
  • Full request audit trail and activity export
  • Organization roles and shared visibility
  • Dedicated onboarding support

Claude API is in preview: no live API requests are processed yet. An eligible subscription opens the preview workspace.

Pricing FAQ

Compare model costs, understand BYOK fees, and see how billing works before you scale traffic.

How should a team estimate spend before switching production traffic?

Start from the Claude model catalog: compare per-1M-token input and output rates for the families you plan to use, then multiply by your expected prompt and completion volume including retries and tool calls.

Does BYOK mean Claude API becomes free?

No. Bring-your-own-key keeps provider token costs on your own provider account; the subscription covers routing policy, usage visibility and support on the Claude API side.

Where can I verify what was actually billed?

Usage and spend are attributed per API key, and billing history stays available in your account settings. Provider invoices remain the source of truth for provider-side charges.

Is the pricing page enough to make a finance decision?

Use it as the starting point for a budget review, then combine it with your own traffic measurements, the usage accounting views and current provider rate cards before you commit production traffic.

Start building in 3 lines of code

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://claudeapi.lat/v1",
  apiKey: "sk-your-key",
});

const response = await client.chat.completions.create({
  model: "anthropic/claude-sonnet-4.6",
  messages: [{ role: "user", content: "Hello!" }],
});