Text Similarity Scorer

Choose this endpoint when you need a multi-metric text similarity comparison that combines edit distance, word overlap, and cosine similarity in a single call with a human-readable verdict. It is ideal for duplicate detection, plagiarism checking, paraphrase detection, or content change monitoring where a nuanced score rather than a binary yes/no is needed. Prefer this over single-metric APIs when you want triangulated confidence across three distinct similarity methods.

• Missing required query parameters 'a' or 'b' returns an error
• Input text exceeding 1500 characters for field 'a' may be truncated or rejected
• Empty string inputs may produce degenerate similarity scores
• Very long texts for parameter 'b' may degrade performance or accuracy
• Network or upstream service errors returning non-200 responses
• Payment failure via x402 protocol preventing request from completing