LLMR-package: LLMR: unified LLM workflows for R

In LLMR: Interface for Large Language Model APIs in R

LLMR: unified LLM workflows for R

Description

LLMR provides provider-agnostic text generation, embeddings, structured JSON, XML-like tag extraction, chat sessions, robust retries, and parallel experiment utilities.

Common workflows

• One prompt: call_llm() or call_llm_robust()

• Vectors or data frames: llm_fn()

• dplyr pipelines: llm_mutate(), including shorthand llm_mutate(new_col = "{existing_col}", .config = cfg)

• Strict structured fields: llm_mutate_structured() or llm_mutate() with .structured = TRUE

• Soft structured fields: llm_mutate_tags() or llm_mutate() with .tags

• Factorial experiments: build_factorial_experiments() and call_llm_par()

Reliability and parallel execution

call_llm_robust() retries transient failures. setup_llm_parallel() controls worker count for parallel helpers; llm_fn() and llm_mutate() use call_llm_broadcast() internally.

Author(s)

Maintainer: Ali Sanaei [e-mail]

See Also

Useful links:

• https://github.com/asanaei/LLMR

• https://asanaei.github.io/LLMR/

• Report bugs at https://github.com/asanaei/LLMR/issues

LLMR documentation built on May 22, 2026, 1:07 a.m.