Benchmarks
Every suite, every competitor
Results as published in the repository README. Each suite links to its detail page with heap measurements and the exact command to reproduce it on your machine.
Validation
Boolean guard over a realistic user object.
Structured issue reporting on a deeply broken input.
High-load validation
Preallocated batch of unknown users, validation work only.
Same benchmark at 100k rows — jit stays allocation-free.
Structured parsing over a preallocated batch.
Issue collection at the end of a 100k batch.
Data operations
Schema-aware deep equality over a large collection.
Structural diff entries over nested collections.
Surgical immutable update of a nested object (no Proxy).
Serialization & streaming
Compiled serializer with static keys and escape fast path.
Progressive validation fails fast while the payload streams in.
Binary rowsets
Byte-offset scan over columnar ArrayBuffer rows.
Aggregation with columnBase + i hot loops.
Unknown input → validated rowset → byte-offset query.
End-to-end flows
Validate 50k unknown objects, filter/project admins, serialize the result.
Numbers vary by hardware — treat relative ratios, not absolutes, as the signal.
How these numbers are produced →