Skip to content
HALOWERK

Textwerk

semantic diff

Both versions are split into sentences (Unicode segmentation plus CLDR abbreviations). Identical sentences are aligned first by a longest common subsequence (Myers 1986); identical sentences outside that sequence count as moved. The remaining sentences are embedded with intfloat/multilingual-e5-small (MIT, "query: " prefix for symmetric similarity, mean pooling, pinned int8 ONNX, SHA-256 checked) and paired greedily by highest cosine similarity at or above min_similarity (default 0.9)..

POST https://text.halowerk.com/v1/semantic-diff

Inputs

Fields of the published input schema
FieldTypeRequiredDescription
old_textstringYes
new_textstringYes
min_similaritynumberNoCosine at or above which two different sentences count as versions of each other.
Show full input schema
{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "old_text",
    "new_text"
  ],
  "properties": {
    "old_text": {
      "type": "string",
      "minLength": 1,
      "maxLength": 60000
    },
    "new_text": {
      "type": "string",
      "minLength": 1,
      "maxLength": 60000
    },
    "min_similarity": {
      "type": "number",
      "minimum": 0.5,
      "maximum": 0.99,
      "default": 0.9,
      "description": "Cosine at or above which two different sentences count as versions of each other."
    }
  }
}

Open original schema

Published input example

{
  "old_text": "Die Laufzeit beträgt zwölf Monate. Der Preis beträgt 49 Euro im Monat.",
  "new_text": "Die Laufzeit beträgt zwölf Monate. Der Preis beträgt 59 Euro im Monat. Gerichtsstand ist Berlin."
}

Result & documentation

Machine-readable documentation of this hub

Source of this page: the published HALOWERK catalog. Updated: 2026-11-17 UTC.

More endpoints in this hub

All endpoints in this hub →