llm_mutate

A convenience wrapper around llm_fn that inserts the result as a new
column via <a href="https://dplyr.tidyverse.org/reference/mutate.html">mutate</a>.

Provides a unified interface to interact with multiple Large Language Model (LLM) APIs. The package supports text generation, embeddings, parallelization, as well as tidyverse integration. Users can switch between different LLM providers seamlessly within R workflows, or call multiple models in parallel. The package enables creation of LLM agents for automated tasks and provides consistent error handling across all supported APIs. APIs include 'OpenAI' (see <https://platform.openai.com/docs/overview> for details), 'Anthropic' (see <https://docs.anthropic.com/en/api/getting-started> for details), 'Groq' (see <https://console.groq.com/docs/api-reference> for details), 'Together AI' (see <https://docs.together.ai/docs/quickstart> for details), 'DeepSeek' (see <https://api-docs.deepseek.com> for details), 'Gemini' (see <https://aistudio.google.com> for details), and 'Voyage AI' (see <https://docs.voyageai.com/docs/introduction> for details).

Ali Sanaei

LLMR

Interface for Large Language Model APIs in R

llm_mutate function

<dl><dt>.data</dt>
<dd>A data frame / tibble.</dd>
<dt>output</dt>
<dd>Unquoted name of the new column you want to add.</dd>
<dt>prompt</dt>
<dd>A glue template string.
*If* <code>x</code> is a data frame, use <code>{col}</code> placeholders;
*if* <code>x</code> is a vector, refer to the element as <code>{x}</code>.</dd>
<dt>.config</dt>
<dd>An llm_config object.</dd>
<dt>.system_prompt</dt>
<dd>Optional system message (character scalar).</dd>
<dt>.before, .after</dt>
<dd>Standard <a href="https://dplyr.tidyverse.org/reference/mutate.html">mutate</a> column-placement helpers.</dd>
<dt>...</dt>
<dd>Passed unchanged to call_llm_broadcast (e.g.\ <code>tries</code>,
<code>progress</code>, <code>verbose</code>).</dd></dl>

Arguments

A convenience wrapper around llm_fn that inserts the result as a new
column via <a href='https://dplyr.tidyverse.org/reference/mutate.html'>mutate</a>.

Mutate a data frame with LLM output — llm_mutate

<dl>

<dt>.data</dt>
<dd>A data frame / tibble.</dd>


<dt>output</dt>
<dd>Unquoted name of the new column you want to add.</dd>


<dt>prompt</dt>
<dd>A glue template string.
*If* <code>x</code> is a data frame, use <code>{col}</code> placeholders;
*if* <code>x</code> is a vector, refer to the element as <code>{x}</code>.</dd>


<dt>.config</dt>
<dd>An llm_config object.</dd>


<dt>.system_prompt</dt>
<dd>Optional system message (character scalar).</dd>


<dt>.before, .after</dt>
<dd>Standard <a href='https://dplyr.tidyverse.org/reference/mutate.html'>mutate</a> column-placement helpers.</dd>


<dt>...</dt>
<dd>Passed unchanged to call_llm_broadcast (e.g.\ <code>tries</code>,
<code>progress</code>, <code>verbose</code>).</dd>

</dl>

llm_mutate: Mutate a data frame with LLM output

Description

Usage

Arguments

Details

See Also

Examples