Bun 1.3 / runtime

A dated reading of what is claimed, reported, and independently verified in the current evidence.

Current reading

Readiness band and full integer triple

AI-assisted assembly · derived results

Readiness band

Watch

Primary summary from verified readiness

Evidence strength Growing

Decision

Agent-assisted assessment · independently reviewed · 2026-10-02

Consider Bun 1.3 for selected production services if the exact patch and workload pass compatibility and endurance tests.

Why: Production use is credible, but residual memory leaks, reported Windows embedding crashes and experimental HTTP/3 require operational controls.
Next: Pin the patch, test dependencies and multi-day memory behavior including disconnects, and monitor RSS with graceful recycling. Evaluate a 1.4 migration separately; confirm compiled binaries contain the intended runtime.

Constraints

Native memory growth in HTTP-client and Node-adapter paths can require profiling, transport changes and graceful process recycling.
Node compatibility and native-addon behavior require testing against the application's actual dependencies and platforms.
HTTP/3 in Bun 1.3.14 is explicitly experimental and unsuitable for production deployment.
Reported Windows crash cases involving embedded 1.3.14 binaries can require a rebuilt application; upgrading standalone Bun does not update its embedded runtime.
Short, single-host benchmarks do not predict prolonged memory behavior or performance with real application I/O.

Evidence summary

AI-assisted assembly

Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.

Publication record

Revision 4