Generate training pairs

curl --request POST \ --url https://flow.seekr.com/v1/flow/alignment/generate \ --header 'Authorization: <api-key>' \ --header 'Content-Type: application/json' \ --data ' { "instructions": "<string>", "files": [ "<string>" ], "system_prompt": "<string>", "type": "principle", "vector_database_id": "<string>" } '

{ "id": "<string>", "created_at": "2023-11-07T05:31:56Z", "updated_at": "2023-11-07T05:31:56Z", "completed_at": "2023-11-07T05:31:56Z", "current_step": "<string>", "step_progress": 123, "progress": 123, "estimated_completion_at": "2023-11-07T05:31:56Z", "eta_minutes": 123, "status_message": "<string>" }

Authorizations

Authorization

string

header

required

Seekr API Key without 'Bearer' Prefix

Body

application/json

instructions

string

required

Task description/instructions for the alignment task

files

string[]

required

List of file ids to use for alignment

system_prompt

string | null

Optional system prompt baked into the generated dataset's system role on every conversation. If omitted, one is generated from instructions and the recursive document summary.

type

enum<string>

default:principle

Type of alignment task (principle or context_data)

Available options:

principle,

context_data

vector_database_id

string | null

Optional vector database id to use for context retrieval during context_data alignment

Response

Successful Response

string

required

created_at

string<date-time>

required

status

enum<string>

required

Available options:

pending,

queued,

running,

cancelled,

failed,

completed,

stopped

updated_at

string<date-time> | null

completed_at

string<date-time> | null

current_step

string | null

step_progress

progress

number | null

estimated_completion_at

string<date-time> | null

eta_minutes

integer | null

status_message

string | null