Swift Compiler vs in Pipeline Benchmark

Measured with: swiftc -typecheck on a single file when there is no local Package.swift; when there is a package, scripts/swiftc-bench-typecheck.sh (same Sources + Generated inputs and Clang flags idea as in-cli sil_emit) after a timed swift build. Also package-context swift build (SwiftPM reference) and in build default (native hybrid pipeline only, no SwiftPM). in column = inauguration compile path (scheduler + SIL passes today); swift build = legacy SwiftPM baseline until native codegen fully replaces it. hybrid-cli matches the native wave harness without the in CLI wrapper overhead. Wall times: median over 3 timed runs; min–max across those runs shown in parentheses next to medians (easy tables) or inline (detail table).

Benchmark Environment

Generated (UTC): 2026-05-09T14:19:19Z
Host OS: Darwin
Kernel: 25.4.0
CPU: Apple M5 Pro
Memory: 51539607552
Swift: swift-driver version: 1.148.6 Apple Swift version 6.3 (swiftlang-6.3.0.123.5 clang-2100.0.123.102) Target: arm64-apple-macosx26.0
Rustc: rustc 1.94.1 (e408947bf 2026-03-25)
Cargo: cargo 1.94.1 (29ea6fb6a 2026-03-24)
V: V 0.5.0 e2f5d6c
BENCH_RUNS: 3
BENCH_WARMUP_RUNS: 1
in binary: /Users/undivisible/projects/inauguration/in-cli/target/debug/in
hybrid-cli binary: /Users/undivisible/projects/inauguration/compiler/rust-driver/target/debug/hybrid-cli

Easy Copy/Paste

Example	SwiftPM swift build median (min–max ms)	in native median (min–max ms)
`aurorality/examples/counter`	984.37 (952.47–1004.23)	548.29 (540.68–578.89)
`aurorality/examples/basic`	943.56 (936.91–1000.56)	559.43 (541.97–608.84)
`aurorality/examples/hyperchat`	433.33 (386.77–614.14)	4334.97 (3155.60–4571.39)

SwiftPM Package (Preview Host) Benchmark

Example	SwiftPM swift build median (min–max ms)	in native median (min–max ms)
`runtime/swift-preview-host`	681.30 (647.08–781.19)	771.99 (725.52–774.55)

Example	swiftc med (min–max)	SwiftPM med (min–max)	in native med (min–max)	hybrid-cli med (min–max)	native÷SwiftPM	in-stage-total(ms)	in-driver-overhead(ms)	in-wrapper-overhead(ms)	loss bucket	swift build ok	in ok
`aurorality/examples/counter`	395.13 (395.11–414.95)	984.37 (952.47–1004.23)	548.29 (540.68–578.89)	7.05 (6.87–7.17)	0.557	538.847	9.447	541.248	win	✅	✅
`aurorality/examples/basic`	402.42 (371.62–412.28)	943.56 (936.91–1000.56)	559.43 (541.97–608.84)	6.84 (6.06–7.15)	0.593	550.636	8.792	552.593	win	✅	✅
`aurorality/examples/hyperchat`	2326.91 (1966.04–3817.95)	433.33 (386.77–614.14)	4334.97 (3155.60–4571.39)	15.00 (8.19–15.83)	10.004	4319.776	15.194	4319.970	swift-frontend-stage	✅	✅