{"query":"Benchmark Versioning","corpusVersion":"local","generatedAt":"2026-09-13T04:37:14.722Z","results":[{"blockId":"governance.benchmark-versioning#benchmark-versioning","pageId":"governance.benchmark-versioning","title":"Benchmark Versioning","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#benchmark-versioning","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"benchmark-versioning","kind":"reference","productArea":"governance","score":910.2506511511011,"reasons":["search_match","title_match","display_title_match","term_match"],"markdown":"# Benchmark Versioning"},{"blockId":"governance.benchmark-versioning#related-task-pages","pageId":"governance.benchmark-versioning","title":"Related task pages","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#related-task-pages","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"related-task-pages","kind":"reference","productArea":"governance","score":261.03331626157615,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Related task pages\n\n{% related-card-grid title=\"Related task pages\" %}\n- [Benchmark snapshots](/docs/coverage-engineering/benchmark-snapshots)\n- [Benchmarks](/docs/object-model/benchmarks)\n- [Compare Harness Versions](/docs/benchmark-evaluations/compare)\n- [Product quickstart](/docs/quickstart)\n- [Task index](/docs/operating-manual/task-index)\n{% /related-card-grid %}"},{"blockId":"governance.benchmark-versioning#related-objects","pageId":"governance.benchmark-versioning","title":"Related objects","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#related-objects","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"related-objects","kind":"reference","productArea":"governance","score":249.3802798820098,"reasons":["search_match","page_title_match","term_match"],"markdown":"## Related objects\n\nUse [Dataset Snapshots](/docs/benchmark-datasets/snapshots) to create the frozen dataset boundary. Use [Benchmark Evaluations](/docs/benchmark-evaluations) to inspect Runs inside one exact version, and [Compare Harness Versions](/docs/benchmark-evaluations/compare) to interpret candidate movement without hiding version changes.\n\n{% example-demo title=\"Separating candidate change from standard change\" %}\nHarness Version 12 improves retrieval and is evaluated against Benchmark Version 4, the same boundary used for Version 11. That comparison isolates candidate behavior. Later, experts approve a stricter source-authority Rubric and the dataset gains conflict Cases. The team creates Benchmark Version 5 and reports subsequent Runs under that new boundary instead of presenting the lower score as a regression against unchanged evidence.\n{% /example-demo %}"},{"blockId":"governance.benchmark-versioning#fields-states-or-lifecycle-rules","pageId":"governance.benchmark-versioning","title":"Fields, states, or lifecycle rules","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#fields-states-or-lifecycle-rules","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"fields-states-or-lifecycle-rules","kind":"reference","productArea":"governance","score":249.17179970442845,"reasons":["search_match","page_title_match","term_match"],"markdown":"## Fields, states, or lifecycle rules\n\n- Create a new Snapshot and resulting Benchmark Version when changed evidence would alter what a Run claims to test.\n- Existing Runs remain attached to their original Benchmark Version.\n- Current Dataset edits do not mutate a historical version.\n- A new Harness Version alone does not require a new Benchmark Version; candidate and evidence versions move independently.\n- Comparisons within one Benchmark Version isolate candidate differences more cleanly.\n- Cross-version comparisons must name the changed Cases, evaluators, or representation boundary.\n- Version identity does not prove that coverage is sufficient or that every admitted Rubric is correct."},{"blockId":"governance.benchmark-versioning#source-confidence","pageId":"governance.benchmark-versioning","title":"Source confidence","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#source-confidence","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"source-confidence","kind":"reference","productArea":"governance","score":249.050185331693,"reasons":["search_match","page_title_match","term_match"],"markdown":"## Source confidence\n\nCode-backed: Benchmark Datasets → Snapshots preserves immutable Dataset boundaries, and the version-scoped Evaluation route binds Runs to one selected Benchmark Version. Coverage quality and downstream decisions remain outside version identity itself."},{"blockId":"governance.benchmark-versioning#definition","pageId":"governance.benchmark-versioning","title":"Definition","pageTitle":"Benchmark Versioning","url":"https://teammately.ai/docs/governance/benchmark-versioning.md","humanUrl":"https://teammately.ai/docs/governance/benchmark-versioning#definition","markdownUrl":"https://teammately.ai/docs/governance/benchmark-versioning.md","sectionId":"definition","kind":"reference","productArea":"governance","score":248.87615790369043,"reasons":["search_match","page_title_match","term_match"],"markdown":"## Definition\n\nA Benchmark Version is the immutable evidence boundary used by Runs. It identifies the frozen dataset state and admitted evaluator relationships that make a result interpretable. The Benchmark remains a durable program; its versions preserve successive evidence boundaries as Cases, materials, coverage, Policies, or Rubrics change."},{"blockId":"governance.case-versioning#case-versioning","pageId":"governance.case-versioning","title":"Case Versioning","pageTitle":"Case Versioning","url":"https://teammately.ai/docs/governance/case-versioning.md","humanUrl":"https://teammately.ai/docs/governance/case-versioning#case-versioning","markdownUrl":"https://teammately.ai/docs/governance/case-versioning.md","sectionId":"case-versioning","kind":"reference","productArea":"governance","score":192.6858225060529,"reasons":["search_match","term_match"],"markdown":"# Case Versioning"},{"blockId":"governance.policy-versioning#policy-versioning","pageId":"governance.policy-versioning","title":"Policy Versioning","pageTitle":"Policy Versioning","url":"https://teammately.ai/docs/governance/policy-versioning.md","humanUrl":"https://teammately.ai/docs/governance/policy-versioning#policy-versioning","markdownUrl":"https://teammately.ai/docs/governance/policy-versioning.md","sectionId":"policy-versioning","kind":"reference","productArea":"governance","score":192.6858225060529,"reasons":["search_match","term_match"],"markdown":"# Policy Versioning"}]}