Skip to content

Create Dataset

POST
/v1/datasets
curl --request POST \
--url https://example.com/v1/datasets \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: multipart/form-data' \
--form file=example \
--form source=example \
--form connection_id=example

Ingest a CSV, TSV, JSON, JSONL or Parquet file, optionally gzipped: either upload file or give a source. The format is detected from the file’s content.

Media typemultipart/form-data
Body_createDataset
object
file
Any of:
string
source
Any of:
string
connection_id
Any of:
string

Successful Response

Media typeapplication/json
Dataset
object
id
required
Id
string
name
required
Name
string
source
required
Source
string
row_count
required
Row Count
integer
size_bytes
required
Size Bytes
integer
columns
required
Columns
Array<object>
Column
object
name
required
Name
string
type
required
Type
string
created_at
required
Created At
string format: date-time
parent_id
required
Any of:
string
branch
required
Any of:
string
run
required
Any of:
DatasetRun

The run that makes a dataset.

object
id
required
Id
string
status
required
Status
string
Allowed values: queued running stopping stopped done failed
failed
required
Failed

Answers that failed, which resuming asks again.

integer
done
required
Done

Model calls answered so far.

integer
total
required
Total

Model calls the run makes.

integer
error
required
Any of:
string

Example

{
"run": {
"status": "queued"
}
}

Validation Error

Media typeapplication/json
HTTPValidationError
object
detail
Detail
Array<object>
ValidationError
object
loc
required
Location
Array
msg
required
Message
string
type
required
Error Type
string
input
Input
ctx
Context
object

Example generated

{
"detail": [
{
"loc": [
"example"
],
"msg": "example",
"type": "example",
"input": "example",
"ctx": {}
}
]
}