Condense Tokens
Condense long text sequences of natural language tokens into soft tokens.
POST /47/compress/text
Required attributes
- Name
text- Type
- string
- Description
Text to be compressed
Default:
In machine learning, the perceptron is an algorithm for supervised learning of binary classifiers. A binary classifier is a function which can decide whether or not an input, represented by a vector of numbers, belongs to some specific class. It is a type of linear classifier, i.e. a classification algorithm that makes its predictions based on a linear predictor function combining a set of weights with the feature vector.
Optional attributes
- Name
ooc_status_code- Type
- string
- Description
Status code to return when user runs out of credits (208 or 402)
- Name
top_node_performance- Type
- number
- Description
Performance parameter for compression
Default:
0.1
Request
curl -X POST "https://api.bitmind.ai/oracle/v1/47/compress/text" \
-H "Authorization: Bearer {token}" \
-H "Content-Type: application/json" \
-d '{
"ooc_status_code": "\"example-ooc_status_code\"",
"text": "In machine learning, the perceptron is an algorithm for supervised learning of binary classifiers. A binary classifier is a function which can decide whether or not an input, represented by a vector of numbers, belongs to some specific class. It is a type of linear classifier, i.e. a classification algorithm that makes its predictions based on a linear predictor function combining a set of weights with the feature vector.",
"top_node_performance": 0.1
}'
Response
{
"compressed": "perceptron utilized in machine learning is a supervised learning algorithm designed for binary classification. involves a function that categorizes a numerical input vector to determine its membership in a particular class. As a linear classifier it predicts based on a linear predictor function combines a feature vector with set of weights.",
"metadata": {
"original_tkn_count": 80,
"compressed_tkn_count": 62,
"credit_cost": 9.000000000000001e-7,
"condensed_pct": 22.5
}
}
POST /47/compress/messages
Required attributes
- Name
messages- Type
- array
- Description
List of messages where each message has a 'role' and 'content' key.
Optional attributes
- Name
ooc_status_code- Type
- string
- Description
Status code to return when user runs out of credits (208 or 402)
- Name
compress_user- Type
- boolean
- Description
Whether to compress user messages
Default:
true
- Name
compress_assistant- Type
- boolean
- Description
Whether to compress assistant messages
Default:
true
- Name
top_node_performance- Type
- number
- Description
Performance parameter for compression
Default:
0.1
Request
curl -X POST "https://api.bitmind.ai/oracle/v1/47/compress/messages" \
-H "Authorization: Bearer {token}" \
-H "Content-Type: application/json" \
-d '{
"ooc_status_code": "\"example-ooc_status_code\"",
"messages": [],
"compress_user": true,
"compress_assistant": true,
"top_node_performance": 0.1
}'
Response
{
"compressed": [
{
"role": "user",
"content": "Inform the user suitable message content is required."
}
],
"metadata": {
"original_tkn_count": 19,
"compressed_tkn_count": 16,
"credit_cost": 1.5000000000000002e-7,
"condensed_pct": 15.79
}
}
Last updated: April 24, 2025