Skip to main content
When status is completed or partial and a customer report exists, read it from GET /v1/research/{run_id}/result.
detail defaults to full. Compact is for another model that will read the result. The caller already has the prompt, and a long question spends context. Answer, evidence, analysis, the research record, and methodology stay as stored. Compact does not start research, call a model, or charge credits. Any other detail value is 400 invalid_request. A run that is still queued, running, or synthesizing returns 409 result_not_ready. A finished run with no customer report returns 409 research_failed. A partial run that already has a customer report returns that report. Another workspace’s run is 404 run_not_found. The report does not include original model text. Use model responses for one stored answer, or the full-panel export to download every response.

​
Where to read the finding

Read the finding from the JSON. Do not scrape HTML.
  • report.answer.headline
  • report.answer.direct_answer
  • report.answer.rationale
  • report.answer.key_points
  • report.answer.recommendation
  • report.answer.ranked_options
  • report.evidence.claims
  • report.evidence.camps (the positions)
  • report.analysis.disagreements
  • report.analysis.minority_insights
  • report.analysis.uncertainties
  • report.research_record.provenance (model, lab, and inclusion)
  • report.methodology
research_record and methodology are the stored objects. Methodology is the last field in a compact report. A minority finding lists its models at report.analysis.minority_insights[].supporting_model_ids. Use one of those ids to read that model’s stored text.