File
The `File` object represents a document that has been uploaded.
The File object represents a document that has been uploaded.
| Property | Type | Required | Description |
|---|---|---|---|
id | string | Yes | The file identifier, which can be referenced in the API endpoints. |
project_id | string | Yes | The project identifier file belongs to. |
object | enum | No | The object type, which is always file. One of: file, list. Default: "file". |
bytes | integer | null | No | The size of the file, in bytes. Default: 0. |
filename | string | Yes | The name of the file. |
purpose | Purpose | No | The intended purpose of the file. Default: "assistants". |
created_at | integer | No | The Unix timestamp (in seconds) for when the file was created. |
expires_at | integer | No | The Unix timestamp (in seconds) for when the file will expire. |
graph_ingestion_status | GraphIngestionStatus | null | No | Status of the file's most recent graph-ingestion job, or null if never ingested. |
graph_ingestion_job_id | string | null | No | ID of the graph-ingestion job that last wrote graph_ingestion_status. |
extracted_metadata | object | null | No | Metadata fields extracted via the metadata expansion pipeline, keyed by field name. |
summary | string | null | No | Summary generated via the document summarization pipeline. |
content_type | string | null | No | The file's content type, taken from the upload's Content-Type header or guessed from the filename. |
summary_provenance | SummaryProvenance | null | No | How summary was produced: model, options, job, time, and whether the document was truncated. |
We use cookies on our website. We use them to ensure proper functioning of the site and, if you agree, for purposes such as analytics, marketing, and targeting ads.