Self-hosted Docker

docker run -d --name llmmeter \
  -p 8080:8080 \
  -e LLMMETER_DB_URL=postgres://user:pass@db/llmmeter \
  -e LLMMETER_INGEST_TOKEN=$(openssl rand -hex 32) \
  -e LLMMETER_DASHBOARD_TOKEN=$(openssl rand -hex 32) \
  ghcr.io/amit641/llmmeter:latest

The image runs llmmeter serve, which exposes:

• POST /ingest — accepts batched LLMCallRecord[] (Bearer token required)
• /api/* — read API for the dashboard (token-protected)
• / — the embedded React dashboard (token-protected)

Wire your app to the collector

import { meter, httpSink } from "@amit641/llmmeter";

const sink = httpSink({
  url: "https://llmmeter.mycompany.com/ingest",
  token: process.env.LLMMETER_INGEST_TOKEN,
});
const openai = meter(new OpenAI(), { sink });

Storage

• SQLite (default): a single Docker volume is enough for tens of millions of records.
• Postgres (LLMMETER_DB_URL): for multi-instance deployments or when you already have managed Postgres.

Pruning

Schedule a cron:

docker exec llmmeter llmmeter prune --pg "$LLMMETER_DB_URL" --older-than 90d