Small experiment with LLMR

knitr::opts_chunk$set(
  collapse = TRUE, comment = "#>",
  eval = identical(tolower(Sys.getenv("LLMR_RUN_VIGNETTES", "false")), "true") )

Overview

This vignette demonstrates:

1. Building factorial experiment designs with build_factorial_experiments()
2. Running experiments in parallel with call_llm_par()
3. Comparing unstructured vs. structured output across providers

The workflow is: design → parallel execution → analysis

We will compare three configurations on two prompts, once unstructured and once with structured output. In choosing models, note that at the time of writing this vignette, Gemini models are not guaranteeing the schema output and is more likely to run into trouble.

library(LLMR)
library(dplyr)
cfg_openai <- llm_config("openai",   "gpt-5-nano")
cfg_cld    <- llm_config("anthropic","claude-sonnet-4-20250514", max_tokens = 512)
cfg_gem    <- llm_config("groq",     "openai/gpt-oss-20b")

Build a factorial design

experiments <- build_factorial_experiments(
  configs       = list(cfg_openai, cfg_cld, cfg_gem),
  user_prompts  = c("Summarize in one sentence: The Apollo program.",
                    "List two benefits of green tea."),
  system_prompts = c("Be concise.")
)
experiments

Run unstructured

setup_llm_parallel(workers = 10)
res_unstructured <- call_llm_par(experiments, progress = TRUE)
reset_llm_parallel()
res_unstructured |>
  select(provider, model, user_prompt_label, response_text, finish_reason) |>
  head()

Understanding the results:

The finish_reason column shows why each response ended:

• "stop": normal completion
• "length": hit token limit (increase max_tokens)
• "filter": content filter triggered

The user_prompt_label helps track which experimental condition produced each response.

Structured version

schema <- list(
  type = "object",
  properties = list(
    answer = list(type="string"),
    keywords = list(type="array", items = list(type="string"))
  ),
  required = list("answer","keywords"),
  additionalProperties = FALSE
)

experiments2 <- experiments
experiments2$config <- lapply(experiments2$config, enable_structured_output, schema = schema)

setup_llm_parallel(workers = 10)
res_structured <- call_llm_par_structured(experiments2 , .fields = c("answer","keywords") )
reset_llm_parallel()

res_structured |>
  select(provider, model, user_prompt_label, structured_ok, answer) |>
  head()