The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
knitr::opts_chunk$set(
collapse = TRUE, comment = "#>",
eval = identical(tolower(Sys.getenv("LLMR_RUN_VIGNETTES", "false")), "true")
)We’ll show both unstructured and structured pipelines, using four model names: - gpt-5-nano (OpenAI) - claude-sonnet-4-20250514 (Anthropic) - gemini-2.5-flash (Gemini) - openai/gpt-oss-20b (Groq)
You will need environment variables OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY.
library(LLMR)
library(dplyr)
cfg_openai <- llm_config("openai", "gpt-5-nano")
cfg_cld <- llm_config("anthropic","claude-sonnet-4-20250514", max_tokens = 512) # avoid warnings; Anthropic requires max_tokens
cfg_gemini <- llm_config("gemini", "gemini-2.5-flash")
cfg_groq <- llm_config("groq", "openai/gpt-oss-20b")words <- c("excellent", "awful", "fine")
out <- llm_fn(
words,
prompt = "Classify '{x}' as Positive, Negative, or Neutral.",
.config = cfg_openai,
.return = "columns"
)
out
#> # A tibble: 3 × 14
#> response_text finish_reason sent_tokens rec_tokens total_tokens
#> <chr> <chr> <int> <int> <int>
#> 1 Positive stop 19 138 157
#> 2 Negative. stop 20 204 224
#> 3 Neutral stop 19 266 285
#> # ℹ 9 more variables: reasoning_tokens <int>, success <lgl>,
#> # error_message <chr>, status_code <int>, error_code <chr>, bad_param <chr>,
#> # response_id <chr>, duration <dbl>, raw_response_json <chr>out_groq <- llm_fn(
words,
prompt = "Classify '{x}' as Positive, Negative, or Neutral.",
.config = cfg_groq,
.return = "columns"
)
out_groq
#> # A tibble: 3 × 14
#> response_text finish_reason sent_tokens rec_tokens total_tokens
#> <chr> <chr> <int> <int> <int>
#> 1 Positive stop 84 50 134
#> 2 Negative stop 85 89 174
#> 3 Positive stop 84 103 187
#> # ℹ 9 more variables: reasoning_tokens <int>, success <lgl>,
#> # error_message <chr>, status_code <int>, error_code <chr>, bad_param <chr>,
#> # response_id <chr>, duration <dbl>, raw_response_json <chr>schema <- list(
type = "object",
properties = list(
label = list(type = "string", description = "Sentiment label"),
score = list(type = "number", description = "Confidence 0..1")
),
required = list("label", "score"),
additionalProperties = FALSE
)
out_s <- llm_fn_structured(
x = words,
prompt = "Classify '{x}' as Positive, Negative, or Neutral with confidence.",
.config = cfg_openai,
.schema = schema,
.fields = c("label", "score")
)
out_s
#> # A tibble: 3 × 20
#> response_text finish_reason sent_tokens rec_tokens total_tokens
#> <chr> <chr> <int> <int> <int>
#> 1 "{\"label\":\"Positive\",\"… stop 69 409 478
#> 2 "{\"label\":\"Negative\",\"… stop 70 410 480
#> 3 "{\"label\":\"Neutral\",\"s… stop 69 473 542
#> # ℹ 15 more variables: reasoning_tokens <int>, success <lgl>,
#> # error_message <chr>, status_code <int>, error_code <chr>, bad_param <chr>,
#> # response_id <chr>, duration <dbl>, raw_response_json <chr>,
#> # structured_ok <lgl>, structured_data <list>, label <chr>, score <dbl>,
#> # structured_valid <lgl>, structured_error <chr>df <- tibble::tibble(
id = 1:3,
text = c("Cats are great pets", "The weather is bad", "I like tea")
)
df_u <- df |>
llm_mutate(
answer,
prompt = "Give a short category for: {text}",
.config = cfg_cld,
.return = "columns"
)
df_u
#> # A tibble: 3 × 15
#> answer answer_finish answer_sent answer_rec answer_tot answer_reason answer_ok
#> <chr> <chr> <int> <int> <int> <int> <lgl>
#> 1 **Pet… stop 18 14 32 NA TRUE
#> 2 **Wea… stop 17 14 31 NA TRUE
#> 3 **Bev… stop 16 10 26 NA TRUE
#> # ℹ 8 more variables: answer_err <chr>, answer_id <chr>, answer_status <int>,
#> # answer_ecode <chr>, answer_param <chr>, answer_t <dbl>, id <int>,
#> # text <chr>schema2 <- list(
type = "object",
properties = list(
category = list(type = "string"),
rationale = list(type = "string")
),
required = list("category", "rationale"),
additionalProperties = FALSE
)
df_s <- df |>
llm_mutate_structured(
annot,
prompt = "Extract category and a one-sentence rationale for: {text}",
.config = cfg_gemini,
.schema = schema2
# Because a schema is present, fields auto-hoist; you can also pass:
# .fields = c("category", "rationale")
)
df_s
#> # A tibble: 3 × 19
#> annot annot_finish annot_sent annot_rec annot_tot annot_reason annot_ok
#> <chr> <chr> <int> <int> <int> <int> <lgl>
#> 1 "{\n \"cat… stop 15 31 46 923 TRUE
#> 2 "{\n \"cat… stop 15 30 45 129 TRUE
#> 3 "{\n \"cat… stop 14 36 50 129 TRUE
#> # ℹ 12 more variables: annot_err <chr>, annot_id <chr>, annot_status <int>,
#> # annot_ecode <chr>, annot_param <chr>, annot_t <dbl>, id <int>, text <chr>,
#> # structured_ok <lgl>, structured_data <list>, category <chr>,
#> # rationale <chr>These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.