Structured LLM Responses
The "types" feature of the Developer API empowers you to retrieve structured data that is validated against specific data types. With the addition of a "context_id", the API provides live, context-aware responses. Below you will find examples for each data type supported by the /v1/types/<type>
endpoint.
Using the Types Endpoint
Each data type (integer
, boolean
, datetime
, string
, float
, and their respective array types) can be queried through this endpoint. When making a POST request, replace <type>
with the desired data type.
If you aren't passing a context ID, use the content
key:
curl -X POST https://api.gethyper.ai/v1/types/boolean \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"content": "Is the earth flat?"}'
Otherwise, you should use both the query
and context_id
keys as in the examples below.
Integer
curl -X POST https://api.gethyper.ai/v1/types/integer \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "How many employees work at the London office?", "context_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"}'
Response
Example response
{
"data": 125
}
Boolean
curl -X POST https://api.gethyper.ai/v1/types/boolean \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "Is the New York office open on Sundays?", "context_id": "b2c3d4e5-f6a7-890b-cdef-1234567890ab"}'
Response
Example response
{
"data": false
}
Datetime
curl -X POST https://api.gethyper.ai/v1/types/datetime \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "When is the next company-wide town hall meeting?", "context_id": "c3d4e5f6-a7b8-90cd-efab-2345678901cd"}'
Response
Example response
{
"data": "2023-11-15T10:00:00Z"
}
String
curl -X POST https://api.gethyper.ai/v1/types/string \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "Who is the CTO of Hyper?", "context_id": "d4e5f6a7-b8c9-0def-ab12-3456789012de"}'
Response
Example response
{
"data": "Samir Khoja"
}
Float
curl -X POST https://api.gethyper.ai/v1/types/float \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "What was the revenue growth percentage in Q2?", "context_id": "e5f6a7b8-c9d0-efab-1234-5678901234ef"}'
Response
Example response
{
"data": 4.7
}
Integer Array
curl -X POST https://api.gethyper.ai/v1/types/integer_array \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "What is the headcount for each department?", "context_id": "a1b2c3d4-e5f6-4g7h-8i9j-0k1l2m3n4o5p"}'
Response
Example response
{
"data": [25, 40, 15, 50, 30]
}
Boolean Array
curl -X POST https://api.gethyper.ai/v1/types/boolean_array \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "Are services meeting performance targets?", "context_id": "5f6a7b8c-9d0e-1f2a-3b4c-5d6e7f8g9h0i"}'
Response
Example response
{
"data": [true, false, true, true, false]
}
Datetime Array
curl -X POST https://api.gethyper.ai/v1/types/datetime_array \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "What are the upcoming project deadlines?", "context_id": "12345678-90ab-cdef-g123-456789abcdef"}'
Response
Example response
{
"data": ["2023-11-15T17:00:00Z", "2023-12-01T17:00:00Z", "2023-12-20T17:00:00Z"]
}
String Array
curl -X POST https://api.gethyper.ai/v1/types/string_array \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "List all department names", "context_id": "9a8b7c6d-5e4f-3a2b-1c0d-e9f8a7b6c5d4"}'
Response
Example response
{
"data": ["Human Resources", "Finance", "Research and Development", "Sales", "Customer Support"]
}
Float Array
curl -X POST https://api.gethyper.ai/v1/types/float_array \
-H "Authorization: Bearer {your_api_key}" \
-H "Content-Type: application/json" \
-d '{"query": "What were the customer satisfaction ratings from the last survey?", "context_id": "0f9e8d7c-6b5a-4c3d-2e1f-0a9b8c7d6e5f"}'
Response
Example response
{
"data": [4.2, 3.8, 4.5, 4.7, 3.9]
}