Query an LLM — llm • metacheck

Ask a large language model (LLM) any question you want about a vector of text or the text from a text_search(). When type is provided, uses ellmer's structured output API to guarantee output conforming to the type spec; otherwise returns free-text responses in an answer column.

Usage

llm(
  text,
  system_prompt,
  type = NULL,
  text_col = "text",
  model = llm_model(),
  params = list()
)

Arguments

text: The text to send to the LLM (vector of strings, or data frame with the text in a column)
system_prompt: A system prompt to set the behavior of the assistant
type: An optional ellmer type specification for structured extraction (e.g., from type_object(), type_from_schema()). When provided, the provider enforces the schema and returns structured columns instead of free text.
text_col: The name of the text column if text is a data frame
model: the LLM model name (see llm_model_list()) in the format "provider" or "provider/model"
params: a named list to pass to ellmer::params()

Value

a data frame of results

Details

You will need to get your own API key from https://console.groq.com/keys. To avoid having to type it out, add it to the .Renviron file in the following format (you can use usethis::edit_r_environ() to access the .Renviron file)

GROQ_API_KEY="key_value_asdf"

See https://console.groq.com/docs for more information

Examples

if (FALSE) { # \dontrun{
# Free-text query
text <- c("hello", "number", "ten", 12)
system_prompt <- "Is this a number? Answer only 'TRUE' or 'FALSE'"
is_number <- llm(text, system_prompt)

# Structured extraction
type_spec <- ellmer::type_object(
  is_number = ellmer::type_boolean("Whether the input is a number")
)
result <- llm(c("hello", "42"), "Classify the input.", type = type_spec)
} # }