High-level function to generate embeddings for a single text string using OpenAI's embedding API. This function handles the entire process from request creation to response processing.
Usage
oai_embed_text(
text,
model = "text-embedding-3-small",
dimensions = NULL,
max_retries = 5,
timeout = 20,
endpoint_url = "https://api.openai.com/v1/embeddings",
key_name = "OPENAI_API_KEY",
tidy = TRUE
)Arguments
- text
Character string to generate embeddings for (must be non-empty)
- model
OpenAI embedding model to use (default: "text-embedding-3-small")
- dimensions
Number of embedding dimensions (NULL uses model default)
- max_retries
Maximum retry attempts for failed requests (default: 5)
- timeout
Request timeout in seconds (default: 20)
- endpoint_url
OpenAI API endpoint URL (default: OpenAI's embedding endpoint)
- key_name
Name of environment variable containing the API key (default: "OPENAI_API_KEY")
- tidy
Whether to return a tidy tibble format (default: TRUE)
Value
If tidy = TRUE, returns a tibble with embedding vectors as columns
(V1, V2, etc.). If tidy = FALSE, returns the raw httr2 response object.
Details
This function is designed for single text inputs. For processing multiple
texts, use oai_embed_batch() which is more efficient for batch operations.
The function automatically handles API authentication, request retries,
and error handling. By default, it returns a tidy tibble with embedding
vectors as columns, but you can get the raw response by setting tidy = FALSE.
Examples
if (FALSE) { # \dontrun{
# Generate embeddings for a single text
embeddings <- oai_embed_text("Hello world")
# Use a different model with custom dimensions
embeddings <- oai_embed_text(
text = "Hello world",
model = "text-embedding-3-large",
dimensions = 1024
)
# Get raw response instead of tidy format
raw_response <- oai_embed_text(
text = "Hello world",
tidy = FALSE
)
} # }