Data Normalization

Data processing and normalization tasks evaluate structured output from messy prose and different structures. This capability is essential for catalogue and product pipelines where data needs to be extracted from unstructured text and formatted consistently. Models must understand the desired output format and extract relevant information accurately.

Duration

Input Tokens

827

Output Tokens

194

Cost

$0.00

Context

Input

Return if the following product is allowed in EU, USA and Asia. Only follow the guidelines in the text, no other knowledge

Expected output

{
  "allowed_in_eu": false,
  "allowed_in_usa": true,
  "allowed_in_asia": false
}

Model output

{
  "allowed_in_eu": false,
  "allowed_in_usa": true,
  "allowed_in_asia": false
}