Completions
Create completions based on provided messages or prompt and sources.
Completions endpoint is used in scenarios where you need to generate text based on a given context, either provided through a conversation history (messages) or a specific starting prompt. The sources parameter helps to contextualize the completion by specifying the indexes to use.
The response should look something like this:
Body Parameters:
messages
array
An array of message objects. Each message object contains a role and content. This simulates a conversation history.
Yes (or prompt
)
prompt
string
A string prompt to generate completions. This is used when there are no prior messages to base the completion on.
Yes (or messages
)
sources
array
An array of strings representing index ID or DID to be used as a source for generating completions.
Yes
Note: Either messages
or prompt
must be provided, along with sources
.
Example Request Body
Using Messages:
{
"messages": [{
"role": "user",
"content": "Tell me a joke."
}],
"sources": [
"kjzl6kcym7w8y8lefkkbpl44q6jh1zu78gaq2zp18we4wdgz1tz9vxyx213ochh"
]
}
POST /api/discovery/completions HTTP/1.1
Host: dev.index.network
Authorization: Bearer DID_SESSION_SERIALIZED
Content-Type: application/json
Accept: */*
Content-Length: 52
{
"messages": [
{}
],
"prompt": "text",
"sources": [
"text"
]
}
{
"completion": "text"
}
Last updated