zirvDocs

DocsResearch (docs/)

Benchmarks

Indexdocs/benchmarks/View source
FileTitle
build-cost.mdBuild cost: binary size and clean build time
latency-targets.mdLatency targets: hook, dash, wrap
native-runtime-baseline.mdNative-runtime quality baseline
token-cost.mdToken-cost benchmarks
wrapped-vs-vanilla.mdWrapped vs vanilla: what zirv ctx costs and buys

autoresearch#

FileTitle
autoresearch/README.mdautoresearch for zirv (issues #799-#805)
autoresearch/results/2026-09-27-jev-determinism/approve_escalate/report.mdAutoresearch campaign: jev-determinism-approve_escalate
autoresearch/results/2026-09-27-jev-determinism/approve_lower/report.mdAutoresearch campaign: jev-determinism-approve_lower
autoresearch/results/2026-09-27-jev-determinism/classify/report.mdAutoresearch campaign: jev-determinism-classify
autoresearch/results/2026-09-27-jev-determinism/compaction_select/report.mdAutoresearch campaign: jev-determinism-compaction_select
autoresearch/results/2026-09-27-jev-determinism/context/report.mdAutoresearch campaign: jev-determinism-context
autoresearch/results/2026-09-27-jev-determinism/crash/report.mdAutoresearch campaign: jev-determinism-crash
autoresearch/results/2026-09-27-jev-determinism/dispatch/report.mdAutoresearch campaign: jev-determinism-dispatch
autoresearch/results/2026-09-27-jev-determinism/handoff_select/report.mdAutoresearch campaign: jev-determinism-handoff_select
autoresearch/results/2026-09-27-jev-determinism/handoff_thin/report.mdAutoresearch campaign: jev-determinism-handoff_thin
autoresearch/results/2026-09-27-jev-determinism/harvest_screen/report.mdAutoresearch campaign: jev-determinism-harvest_screen
autoresearch/results/2026-09-27-jev-determinism/inject_screen/report.mdAutoresearch campaign: jev-determinism-inject_screen
autoresearch/results/2026-09-27-jev-determinism/inject/report.mdAutoresearch campaign: jev-determinism-inject
autoresearch/results/2026-09-27-jev-determinism/intake_plan/report.mdAutoresearch campaign: jev-determinism-intake_plan
autoresearch/results/2026-09-27-jev-determinism/intake/report.mdAutoresearch campaign: jev-determinism-intake
autoresearch/results/2026-09-27-jev-determinism/judge/report.mdAutoresearch campaign: jev-determinism-judge
autoresearch/results/2026-09-27-jev-determinism/launch_effort/report.mdAutoresearch campaign: jev-determinism-launch_effort
autoresearch/results/2026-09-27-jev-determinism/memory/report.mdAutoresearch campaign: jev-determinism-memory
autoresearch/results/2026-09-27-jev-determinism/missing_tests/report.mdAutoresearch campaign: jev-determinism-missing_tests
autoresearch/results/2026-09-27-jev-determinism/README.mdJev determinism campaigns, 2026-09-27
autoresearch/results/2026-09-27-jev-determinism/review_dedup/report.mdAutoresearch campaign: jev-determinism-review_dedup
autoresearch/results/2026-09-27-jev-determinism/review_disposition/report.mdAutoresearch campaign: jev-determinism-review_disposition
autoresearch/results/2026-09-27-jev-determinism/stop_verify/report.mdAutoresearch campaign: jev-determinism-stop_verify
autoresearch/sample-report.mdAutoresearch campaign: fixture-demo

wrapped-vs-vanilla#

FileTitle
wrapped-vs-vanilla/CONTRACT.mdzirv-vs-vanilla benchmark: shared contract
wrapped-vs-vanilla/quality_rubric.mdWork-quality rubric (blind judge)
wrapped-vs-vanilla/README.mdwrapped-vs-vanilla benchmark harness
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r0/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r1/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r2/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r3/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r3b/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r4-t24/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r4/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r5/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r5b-t24/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r5b/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r6-t24/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r6/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r7-t23/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r7-t24b/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/r7/report.mdHeadline (mean per task, every task weighted equally)
wrapped-vs-vanilla/results/2026-09-24-fix-loop/README.mdBenchmark → fix loop, 2026-09-24/25: results and handover
wrapped-vs-vanilla/tasks/README.mdledgerlite benchmark tasks
wrapped-vs-vanilla/tasks/t05_dedupe/rubric.mdRubric: t05_dedupe
wrapped-vs-vanilla/tasks/t08_usage_doc/rubric.mdRubric: t08_usage_doc
wrapped-vs-vanilla/tasks/t23_afternoon/rubric/step_09.mdStep 9 rubric: wrap-up (README + summary)
wrapped-vs-vanilla/tasks/t24_long_haul/rubric/step_11.mdStep 11 rubric: mid-session design note
wrapped-vs-vanilla/tasks/t24_long_haul/rubric/step_22.mdStep 22 rubric: final wrap-up (README + summary)
wrapped-vs-vanilla/tasks/t24b_long_haul/rubric/step_11.mdStep 11 rubric: mid-session design note
wrapped-vs-vanilla/tasks/t24b_long_haul/rubric/step_22.mdStep 22 rubric: final wrap-up (README + summary)
wrapped-vs-vanilla/tasks/t25_sticky_notes/rubric/step_06.mdStep 6 rubric: wrap-up (docs + summary)
wrapped-vs-vanilla/template/README.mdledgerlite