{"query":"Connect Model Outputs","corpusVersion":"local","generatedAt":"2026-09-13T04:40:04.046Z","results":[{"blockId":"integrations.connect-model-outputs#connect-model-outputs","pageId":"integrations.connect-model-outputs","title":"Connect Model Outputs","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#connect-model-outputs","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"connect-model-outputs","kind":"task","productArea":"benchmark_evaluations","score":2318.0887641447284,"reasons":["search_match","title_match","display_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"# Connect Model Outputs"},{"blockId":"integrations.connect-model-outputs#task-steps-connect-external-model-outputs","pageId":"integrations.connect-model-outputs","title":"Task steps: Connect external model outputs","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#task-steps-connect-external-model-outputs","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"task-steps-connect-external-model-outputs","kind":"task","productArea":"benchmark_evaluations","score":1744.2451716138028,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"### Task steps: Connect external model outputs\n\n1. Open the intended Benchmark Version and go to **Benchmark Evaluations** → **Runs**.\n2. Choose **Import reference outputs** and name the external system or candidate clearly.\n3. Download the mapping template for the current Benchmark Version. Keep `case_id` unchanged; use input and context columns only to verify the match.\n4. Populate `output` for each Case. Add latency, usage, or cost columns only for values measured by the producing system.\n5. Upload the file and inspect unknown Case IDs, missing Benchmark Cases, duplicates, and output previews.\n6. Resolve every mapping error. Do not join on input text or force an output onto a similar-looking Case.\n7. Confirm the mapping and inspect the output-only reference Run.\n8. Wait for evaluation to complete before interpreting result summaries or failures."},{"blockId":"integrations.connect-model-outputs#map-outputs","pageId":"integrations.connect-model-outputs","title":"Map outputs","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#map-outputs","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"map-outputs","kind":"task","productArea":"benchmark_evaluations","score":1197.7523143423507,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Map outputs"},{"blockId":"integrations.connect-model-outputs#related-troubleshooting-pages","pageId":"integrations.connect-model-outputs","title":"Related troubleshooting pages","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#related-troubleshooting-pages","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"related-troubleshooting-pages","kind":"task","productArea":"benchmark_evaluations","score":700.4802690976447,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Related troubleshooting pages\n\n{% related-card-grid title=\"Related troubleshooting pages\" %}\n- [Output mapping](/docs/troubleshooting/output-mapping)\n- [Missing outputs](/docs/troubleshooting/missing-outputs)\n- [Benchmark results changed](/docs/troubleshooting/benchmark-results-changed-unexpectedly)\n{% /related-card-grid %}"},{"blockId":"integrations.connect-model-outputs#object-and-state-changes","pageId":"integrations.connect-model-outputs","title":"Object and state changes","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#object-and-state-changes","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"object-and-state-changes","kind":"task","productArea":"benchmark_evaluations","score":695.4688839110048,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Object and state changes\n\nThe workflow creates an ordinary output-only reference Run for one immutable Benchmark Version and associates submitted outputs with its Cases. It can also store measured telemetry supplied with those outputs.\n\nIt does not create or save a Harness Version, change Benchmark membership, mutate Cases, approve reference responses, or make the external candidate available to Improve as an executable Harness."},{"blockId":"integrations.connect-model-outputs#related-reference-pages","pageId":"integrations.connect-model-outputs","title":"Related reference pages","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#related-reference-pages","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"related-reference-pages","kind":"task","productArea":"benchmark_evaluations","score":695.0063819343561,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Related reference pages\n\n{% related-card-grid title=\"Related reference pages\" %}\n- [Map External Evaluation Outputs](/docs/benchmark-evaluations/output-mapping)\n- [Benchmark Evaluations](/docs/benchmark-evaluations)\n- [Dataset Snapshots](/docs/benchmark-datasets/snapshots)\n- [Integrations](/docs/integrations)\n{% /related-card-grid %}"},{"blockId":"integrations.connect-model-outputs#before-and-after","pageId":"integrations.connect-model-outputs","title":"Before and after","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#before-and-after","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"before-and-after","kind":"task","productArea":"benchmark_evaluations","score":691.037436619303,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Before and after\n\n| before | after |\n| --- | --- |\n| Outputs exist in an external file or system | Outputs are mapped to exact immutable Benchmark Cases |\n| Candidate identity is customer-owned context | A Teammately reference Run preserves both its own ID and the external reference label |\n| No Teammately evaluation state exists | Rubric evaluation proceeds as a separate Run lifecycle |\n| Missing telemetry may be ambiguous | Unmeasured telemetry remains absent rather than becoming zero |"},{"blockId":"integrations.connect-model-outputs#common-failure-modes","pageId":"integrations.connect-model-outputs","title":"Common failure modes","pageTitle":"Connect Model Outputs","url":"https://teammately.ai/docs/integrations/connect-model-outputs.md","humanUrl":"https://teammately.ai/docs/integrations/connect-model-outputs#common-failure-modes","markdownUrl":"https://teammately.ai/docs/integrations/connect-model-outputs.md","sectionId":"common-failure-modes","kind":"task","productArea":"benchmark_evaluations","score":690.7729041348323,"reasons":["search_match","page_title_match","term_match","prefix_or_fuzzy_match"],"markdown":"## Common failure modes\n\n- Reusing IDs from the editable Project Case collection instead of the frozen Benchmark Version.\n- Joining on input text, row order, or a customer ID without verifying the Teammately Case ID.\n- Uploading outputs for two candidate versions under one reference label.\n- Reporting missing latency or cost as zero.\n- Treating successful mapping as successful evaluation.\n- Assuming the reference Run can enter Harness Compare, Arena, or Improve as an executable candidate.\n\n{% example-demo title=\"Retrieval candidate outputs\" %}\nA retrieval team evaluates a new indexing configuration outside Teammately. It exports one answer per frozen Benchmark Case and preserves its own generation ID. In the mapping template, each answer joins on `case_id`; the generation ID remains correlation context and measured latency is included.\n\nThe imported set becomes a reference Run. Teammately evaluates its outputs against the Benchmark Version's Policy and Rubric evidence, while the indexing configuration itself remains outside Teammately as a non-Harness system.\n{% /example-demo %}"}]}