Gemini | Embeddings

Generate vector embeddings for text content. Embeddings capture the semantic meaning of text and can be used for similarity search, clustering, and other NLP tasks.

Simple Example

Generate embeddings for text.

Gemini := TsgcHTTP_API_Gemini.Create(nil);
Gemini.GeminiOptions.ApiKey := 'API_KEY';
WriteLn(Gemini._EmbedContent('text-embedding-004', 'Hello world'));

Advanced Example

Use the typed response classes for full control over embedding data.

Gemini := TsgcHTTP_API_Gemini.Create(nil);
Gemini.GeminiOptions.ApiKey := 'API_KEY';

oEmbedding := Gemini.EmbedContent('text-embedding-004', 'Hello world');
Try
  WriteLn('Dimensions: ' + IntToStr(Length(oEmbedding.Values)));
  for i := 0 to Length(oEmbedding.Values) - 1 do
    WriteLn(FloatToStr(oEmbedding.Values[i]));
Finally
  oEmbedding.Free;
End;