Middleware for the Vercel AI SDK that captures LLM usage, calculates exact costs, and sends normalized billing events directly to your billing platforms.
Whether you’re using OpenAI, Anthropic, OpenRouter, or local models, @ai-billing ensures that every token—including prompt caching and reasoning tokens—is precisely measured and billed to the correct user.
Provider-aware metrics
Understands provider-specific metrics like reasoning tokens, prompt caching, and web search costs so you don’t miss a cent.
Flexible cost calculation
Use native provider pricing (like OpenRouter), hardcoded price maps, or a compatible resolver backed by the Pricing API.
Multiple destinations
Format and forward usage seamlessly to Stripe, Polar, OpenMeter, Lago, or your own endpoint—all at once.
Drop-in middleware
Works with
wrapLanguageModel—no changes to your existing streamText / generateText calls.Install
@ai-billing/openai) and your billing destination (e.g., @ai-billing/stripe). See Provider middleware below.
Quickstart
Start with the AI-billing quickstart to capture usage and provider-reported costs with a console destination. Then add a billing destination and Customer Tagging from authenticated server state. For direct providers that need rates, use a verified static price map or a compatible price resolver. The Pricing API provides public, keyless REST access. The current@ai-billing/narev client still requires an apiKey option; see access and compatibility before choosing that client.
Architecture
The library is built on three composable primitives:1
Provider middlewares
Specialized middleware for each AI SDK provider (e.g.,
@ai-billing/openai, @ai-billing/anthropic). These understand provider-specific metadata shapes, extracting standard tokens alongside advanced metrics like inputCacheReadTokens, internalReasoningTokens, and webSearch.2
Price resolvers
Functions that turn token metrics into precise financial costs. Use
createObjectPriceResolver from @ai-billing/core for custom mappings, or use a compatible live resolver. Skip a separate resolver if your provider (like OpenRouter or Vercel AI Gateway) reports costs natively.3
Destinations
Functions that receive a normalized
BillingEvent and handle the API call to an external billing service (e.g., @ai-billing/stripe, @ai-billing/polar). Configure identity mapping and meter units for your destination. Polar uses base currency units by default; other destinations can use nanos.Packages
Every published@ai-billing package has generated TypeDoc reference docs under /ai-billing/reference/<package>/typedoc/index.
Core and integrations
Provider middleware
Text model support is the current priority. Each package ships V3 and V4 middleware.Billing destinations
Roadmap
Note: We are prioritizing support for text models.Active development To prioritize a provider that is not listed above, open a GitHub issue.
Reference
Core primitives
Middleware, destinations, price resolvers, and shared types in
@ai-billing/core.Narev pricing client
createNarevClient, createNarevPriceResolver, and trace cost helpers in @ai-billing/narev.Next.js components
Chat UI, credit usage widgets, and server factories in
@ai-billing/nextjs.Full API reference
Browse the sidebar for every provider, destination, and integration package.