Get a session details

GEThttps://studioapi.kriyam.ai/models/docwise/{session_id}

Returns a session with each document, its per-stage progress and whatever results are ready. status is completed, partial or failed once the work is over.

Request

curl -X GET "https://studioapi.kriyam.ai/models/docwise/{session_id}" \
  -H "X-API-Key: $KRIYAM_API_KEY"

Response

{8 items"id":"e3a91b47-0c25-4d6f-9b8a-71f4c2..."
"status":"completed"
"capabilities":[2 items0:"parse"
1:"extract"
]
"processing_mode":"fast"
"credits":14
"documents":[1 item0:{6 items"id":"1d7c4f92-8b3a-4e50-a6c1-9f28b5..."
"original_filename":"invoice.pdf"
"status":"completed"
"document_type":"invoice"
"page_count":3
"pages":[1 item0:{2 items"page_number":1
"images":[1 item0:{4 items"serial":"1.1.2"
"type":"image"
"url":"https://storage.example.com/cr..."
"bbox_2d":[4 items0:442
1:10
2:558
3:100
]
}
]
}
]
}
]
"extract_status":"done"
"extraction":{3 items"json_schema":{2 items"type":"object"
"properties":{5 items"invoice_number":{2 items"type":"string"
"description":"The invoice number printed by ..."
}
"invoice_date":{2 items"type":"string"
"description":"Date of issue, as YYYY-MM-DD"
}
"seller_gstin":{2 items"type":"string"
"description":"GSTIN of the seller, not the b..."
}
"total_amount":{2 items"type":"number"
"description":"Grand total payable including ..."
}
"line_items":{3 items"type":"array"
"description":"One entry per billed line"
"items":{2 items"type":"object"
"properties":{2 items"description":{2 items"type":"string"
"description":"What was billed"
}
"amount":{2 items"type":"number"
"description":"Line total before tax"
}
}
}
}
}
}
"extracted_data":{5 items"invoice_number":"INV-2026-0412"
"invoice_date":"2026-09-14"
"seller_gstin":"29ABCDE1234F1Z5"
"total_amount":118000
"line_items":[1 item0:{2 items"description":"Annual platform licence"
"amount":100000
}
]
}
"field_citations":{1 item"invoice_number":{2 items"element_serials":[1 item0:"1.1.3"
]
"document_ids":[1 item0:"1d7c4f92-8b3a-4e50-a6c1-9f28b5..."
]
}
}
}
}

Authentication

Send your key in the X-API-Key header. See Authentication.

Request

Path parameters

Name	Type	Required	Description
`session_id`	string	Required	The `session_id` returned when the session was started.

Response

200 OK

Field	Type	Description
`id`	uuid	The session.
`status`	string	`processing` while work remains; then `completed`, `partial` when some steps or documents finished and others didn't, or `failed` when nothing could be processed.
`capabilities`	string[]	The steps this session actually runs — what you asked for plus anything those depend on.
`processing_mode`	string	`fast` or `accurate`, as sent.
`credits`	integer	Charged so far. It rises as each step delivers, so it is only final once `status` is.
`documents[]`	object[]	Each document with its own status and whatever results are ready.
`documents[].status`	string	This document's own outcome — one can fail while its siblings succeed, which is what `partial` at the session level means.
`documents[].document_type`	string \| null	What `classify` decided this is. Null when classify wasn't run, or when the file holds more than one kind of document.
`documents[].page_count`	integer \| null	Pages found in this document.
`documents[].pages[]`	object[]	Each page, in order, with what was read from it.
`documents[].pages[].page_number`	integer	1-based, within the document.
`documents[].pages[].parsed_ocr`	object[] \| null	The page's elements in reading order — `{type, content, bbox_2d, serial, metadata}`. An image element's `content` embeds its picture as a Markdown image; the same picture is in `images` below as a plain URL.
`documents[].pages[].images[]`	object[]	Every image Docwise cut out of the page — photos, logos, signatures, stamps, figures — one entry per image. Empty when the page has none.
`documents[].pages[].images[].url`	string	A presigned download URL, valid for one hour. Download and store the file if you need to keep it; read the session again for a fresh URL.
`documents[].pages[].images[].serial`	string \| null	The `serial` of the element in `parsed_ocr` this image belongs to.
`documents[].pages[].images[].type`	string \| null	That element's type, e.g. `image`.
`documents[].pages[].images[].bbox_2d`	number[] \| null	Where the image sits on the page, `[x0, y0, x1, y1]` on a 0-1000 scale.
`extract_status`	string	Where extraction has got to: `pending`, `in_progress`, `done` or `failed`, or `not_applicable` when `extract` wasn't requested.
`extraction`	object \| null	The extracted data, once extraction has produced it. One result covers the whole session, since a field can draw on more than one document.
`extraction.json_schema`	object	The schema the values were extracted against — your `extraction_schema` when you sent one, otherwise the one Docwise generated from the documents.
`extraction.extracted_data`	object	The values, shaped by that schema. A field the documents don't answer is null rather than guessed.
`extraction.field_citations`	object \| null	For each top-level field, the documents (`document_ids`) and the parsed elements (`element_serials`) its value was read from.

Errors

Status	Code	When
400	—	No file was sent, more than 5 were, no step was requested, or a requested step is unknown or not currently available. Also when `extraction_schema` is sent without `extract`, isn't a JSON Schema object, or has a field with no `description` — the message names the field.

Every keyed call can also get401402403404429— see Errors.