Li Benchmarks

horner_pure_li greenvalidity pass

Tier 1 · wall_time · N=1M Horner · variant pure_li · OS linux

passpass — perf claims allowed when ratio is green

Source: latest.csv:perf_present

Wall-clock row exists only — no explicit correctness signal yet.

Ratio vs best competitor: 1.167 (116.7% of rust speed) (best in series: rust — Li is never labeled best)
Best competitor value: 0.0007 s

No analytical verify rows in ingest yet. Run tier-1 with --verify so latest.csv exports verify_ulps / verify_within_1ulp.

Problem size: N=1M Horner
Category: micro
Pillar: compiler
Package: lic
Li / catalog oracle (mean ± σ): 0.0006 ± 0.000164 s (n=200) / 0.0008 ± 0.000142 s (n=200) (cpp)
Ratio vs catalog oracle: 0.7500×
Best competitor: rust (0.0007 s)
Li relative speed vs SOTA: 1.167 (1.0 = rust speed)
Validity: validity pass
Threshold: 1.2×
Source: lic/benchmarks/tier1_micro/horner_pure_li

PH ids: PH-5b, PH-7e

Compare oracle: cpp

Relative speed vs best competitor (rust) — SOTA = 1.0, higher is better. Absolute s values are in the table below.

Language comparison
Lang	Mean ± σ	Runs	Unit	Variant	OS
li	0.0006 ± 0.000164	200	s	release	linux
cpp	0.0008 ± 0.000142	200	s	release	linux
rust	0.0007 ± 0.000135	200	s	release	linux
julia	0.0007 ± 0.00013	200	s	release	linux
Metric: wall_time (value = mean of timed runs)