Log a generation
curl --request POST \
--url https://api.honeyhive.ai/generations/log \
--header 'Authorization: <authorization>' \
--header 'Content-Type: application/json' \
--data '{
"feedback": {},
"generation": "<string>",
"ground_truth": "<string>",
"hyperparameters": {},
"inputs": {},
"latency": 123,
"metadata": {},
"model": "<string>",
"prompt": "<string>",
"source": "<string>",
"task": "<string>",
"usage": {},
"user_properties": {},
"version": "<string>"
}'
{
"generation": "<string>",
"generation_id": "<string>",
"stream": true,
"version": "<string>"
}
Authorizations
Bearer authentication header of the form Bearer <token>
, where <token>
is your auth token.
Body
The feedback associated with this generation
The text that was generated by the model provider
The ground truth for the generation
The hyperparameters that were used to generate the text
The inputs filled into the prompt
The latency of the generation in milliseconds
The metadata to track for the generation
The model that was used to generate the text
The prompt used to generate the text
The source of the generation
The task for which the generation is being logged
The token usage for the generation
The user properties to track for the generation
The version of the prompt
Response
The generated completion
The unique ID of the generation
Is stream output
The unique ID of the prompt
curl --request POST \
--url https://api.honeyhive.ai/generations/log \
--header 'Authorization: <authorization>' \
--header 'Content-Type: application/json' \
--data '{
"feedback": {},
"generation": "<string>",
"ground_truth": "<string>",
"hyperparameters": {},
"inputs": {},
"latency": 123,
"metadata": {},
"model": "<string>",
"prompt": "<string>",
"source": "<string>",
"task": "<string>",
"usage": {},
"user_properties": {},
"version": "<string>"
}'
{
"generation": "<string>",
"generation_id": "<string>",
"stream": true,
"version": "<string>"
}