embed_text: Generate Embeddings

In rollama: Communicate with 'Ollama' to Run Large Language Models Locally

Generate Embeddings

Description

Generate Embeddings

Usage

embed_text(
  text,
  model = NULL,
  server = NULL,
  model_params = NULL,
  verbose = getOption("rollama_verbose", default = interactive())
)

Arguments

text	text vector to generate embeddings for.
model	which model to use. See https://ollama.com/library for options. Default is "llama3.1". Set option(rollama_model = "modelname") to change default for the current session. See pull_model for more details.
server	URL to one or several Ollama servers (not the API). Defaults to "http://localhost:11434".
model_params	a named list of additional model parameters listed in the documentation for the Modelfile.
verbose	Whether to print status messages to the Console (TRUE/FALSE). The default is to have status messages in interactive sessions. Can be changed with options(rollama_verbose = FALSE).

Value

a tibble with embeddings.

Examples

## Not run: 
embed_text(c(
  "Here is an article about llamas...",
  "R is a language and environment for statistical computing and graphics."))

## End(Not run)

rollama documentation built on April 3, 2025, 7:48 p.m.