Comparison
How Logixlysia compares to evlog and other Elysia loggers
Feature matrix
| Capability | Logixlysia | evlog | @bogeychan/elysia-logger |
|---|---|---|---|
| Elysia plugin | Yes | Yes (evlog/elysia) |
Yes |
| Pino-backed | Yes | Optional drains | Yes |
| Request context accumulation | mergeContext, useLogger() |
log.set(), useLogger() |
Manual |
| Single access log per request | Auto | log.emit() |
Auto |
| File rotation | Yes | Via drains | Limited |
autoRedact PII |
Yes | Varies | Manual |
| WebSocket lifecycle logs | wrapWs |
Manual | Manual |
| OTel trace correlation | logixlysia/otel |
Built-in integrations | Manual |
| AI usage on access log | logixlysia/ai |
evlog/ai |
Manual |
Presets (dev / prod) |
Yes | N/A | N/A |
When to choose Logixlysia
- You want Pino performance with Elysia-native
store.loggerergonomics. - You need file logging and rotation without wiring a separate drain.
- You prefer incremental context (
mergeContext) plus automatic access logs instead of migrating to a wide-event-only API.
When to consider evlog
- You want a unified wide-event pipeline across many frameworks with managed drains (Axiom, etc.).
- You rely heavily on
evlog/aiandevlogauth integrations out of the box.
Benchmarks
Run the monorepo benchmark package:
bun run benchThe benchmark package has three request-path suites: an overhead floor (all sinks disabled — measures plugin dispatch cost only), a structured sink run (no-op transport/drain — measures log assembly and dispatch), and a file sink run. Compare like with like: the floor numbers contain no logging work by design. Results vary by machine — use them for regression tracking, not absolute rankings.