Get Best Available Embeddings

Auto-detects and uses the best available embedding provider with the following priority:

1. Ollama (free, local, fast) - if running

2. sentence-transformers (local Python) - if Python environment is set up

3. OpenAI API - if OPENAI_API_KEY is set

4. Gemini API - if GEMINI_API_KEY is set

Usage

get_best_embeddings(
  texts,
  provider = "auto",
  model = NULL,
  api_key = NULL,
  verbose = TRUE
)

Arguments

texts

Character vector of texts to embed

provider

Character string: "auto" (default), "ollama", "sentence-transformers", "openai", or "gemini". Use "auto" for automatic detection.

model

Character string specifying the embedding model. If NULL, uses default model for the selected provider.

api_key

Optional API key for OpenAI or Gemini providers. If NULL, falls back to environment variables (OPENAI_API_KEY, GEMINI_API_KEY).

verbose

Logical, whether to print progress messages (default: TRUE)

Value

Matrix with embeddings (rows = texts, columns = dimensions)

See also

Other ai: call_gemini_chat(), call_llm_api(), call_ollama(), call_openai_chat(), check_ollama(), describe_image(), generate_topic_content(), get_api_embeddings(), get_content_type_prompt(), get_content_type_user_template(), get_recommended_ollama_model(), list_ollama_models(), run_rag_search()

Examples

if (FALSE) { # \dontrun{
data(SpecialEduTech)
texts <- SpecialEduTech$abstract[1:5]

# Auto-detect best available provider
embeddings <- get_best_embeddings(texts)

# Force specific provider
embeddings <- get_best_embeddings(texts, provider = "ollama")

dim(embeddings)
} # }