Get Job Results | Sarvam API Docs

{
  "type": "extract",
  "annotations": {
    "insured_name": {
      "confidence": 0.95,
      "sources": [
        {
          "document_id": "019fb7da-5106-7839-adc6-4245e437dfe6",
          "filename": "policy.pdf",
          "page_num": 1
        }
      ]
    },
    "policy_number": {
      "confidence": 0.98,
      "sources": [
        {
          "document_id": "019fb7da-5106-7839-adc6-4245e437dfe6",
          "filename": "policy.pdf",
          "page_num": 1
        }
      ]
    },
    "sum_insured": {
      "confidence": 0.92,
      "sources": [
        {
          "document_id": "019fb7da-5106-7839-adc6-4245e437dfe6",
          "filename": "policy.pdf",
          "page_num": 2
        }
      ]
    }
  },
  "job_id": "20260707_6eb74174-e2c3-490d-afe5-d2e7e9e2b6c7",
  "result": {
    "insured_name": "Aditya Chaudhary",
    "policy_number": "ABC-1234",
    "sum_insured": 500000
  },
  "status": "completed",
  "usage": {
    "pages_total": 2,
    "pages_processed": 2,
    "pages_succeeded": 2,
    "pages_failed": 0
  },
  "version": 1
}

Returns terminal job results. Results are never available while a job is running; requesting them before a terminal status returns 409.

Parse results include job_id, type, status, usage and documents.

Extract results include job_id, type, status, usage, result, annotations and version. annotations mirrors the result shape; every leaf has confidence and sources.

Authentication

api-subscription-keystring

API Key authentication via header

Path parameters

job_idstringRequired

The unique identifier of the job.

Query parameters

formatenumOptionalDefaults to json

Output format of the results. Choose md for Markdown, html for HTML, or json for structured JSON. If you omit format, the results are returned as json by default.

Allowed values:

includeenumOptional

Set to details for Classify and Split jobs to add inspected pages, provenance and boundary details. Supporting evidence also needs output.include_evidence on the request.

Allowed values:

Response

extractobject

Job results for the extract pipeline.

parseobject

Job results for the parse pipeline.

classifyobject

Job results for the classify pipeline.

splitobject

Job results for the split pipeline.

Errors

404

Not Found Error

409

Conflict Error