stateless
Last updated: 2026-06-15 22:10 · same-machine, relative comparison.
Throughput / resource usage
| product | version | offered_in/s | consumed_in/s | output_rate | cpu_avg | cpu_p95 | rss_avg_mb | rss_max_mb | window | run |
|---|---|---|---|---|---|---|---|---|---|---|
| turbine | 0.1.0a4 | 750,000 | 462,645 | 149,115 | 106.4 | 110.0 | 1,066.9 | 1,182.2 | 60 | 2026-06-15 22:10 |
| quix | 3.23.6 | - | 188,295 | 60,080 | 100.8 | 103.0 | 211.6 | 214.6 | 60 | 2026-06-14 18:42 |
Analysis (turbine = baseline):
output_rate: 🏆 turbine — quix 0.40× vs turbinecpu_avg(lower better): 🏆 quix — quix 0.95× vs turbinerss_avg_mb(lower better): 🏆 quix — quix 0.20× vs turbine
Sanity — are we benching the same thing?
| product | selectivity (out/in) | expected | breach_rate | enrichment | sample |
|---|---|---|---|---|---|
| turbine | 0.1790 | 0.3100 | 0.196 | ✓ | 500 |
| quix | 0.3037 | - | 0.178 | ✓ | 500 |
Verdict: ✓ same order of magnitude (selectivity & breach_rate within 2× across products).