Build Anthropic Message Batch requests from a tibble of pairs

This helper converts a tibble of writing pairs into a list of Anthropic Message Batch requests. Each request has a unique custom_id of the form "ANTH_<ID1>_vs_<ID2>" and a params object compatible with the /v1/messages API.

Usage

build_anthropic_batch_requests(
  pairs,
  model,
  trait_name,
  trait_description,
  prompt_template = set_prompt_template(),
  reasoning = c("none", "enabled"),
  custom_id_prefix = "ANTH",
  ...
)

Arguments

pairs

Tibble or data frame with at least columns ID1, text1, ID2, text2. Typically created by make_pairs, sample_pairs, and randomize_pair_order.

model

Anthropic Claude model name, for example "claude-sonnet-4-5", "claude-haiku-4-5", or "claude-opus-4-5".

trait_name

Short label for the trait (for example "Overall Quality").

trait_description

Full-text description of the trait or rubric.

prompt_template

Prompt template string, typically from set_prompt_template. The template should embed your full instructions, rubric text, and <BETTER_SAMPLE> tagging convention.

reasoning

Character scalar indicating whether to allow extended thinking; one of "none" or "enabled". See details above.

custom_id_prefix

Prefix for the custom_id field. Defaults to "ANTH" so that IDs take the form "ANTH_<ID1>_vs_<ID2>".

...

Additional Anthropic parameters such as max_tokens, temperature, top_p, or thinking_budget_tokens, which will be passed through to the Messages API.

Value

A tibble with one row per pair and two main columns:

custom_id

Character ID of the form "<PREFIX>_<ID1>_vs_<ID2>".

params

List-column containing the Anthropic Messages API params object for each request, ready to be used in the requests array of /v1/messages/batches.

Details

The function mirrors the behaviour of build_openai_batch_requests but targets Anthropic's /v1/messages/batches endpoint. It applies the same recommended defaults and reasoning constraints as anthropic_compare_pair_live:

• reasoning = "none":

  • Omitted temperature and top_p values are not sent, so the model/provider defaults apply.

  • Default max_tokens = 768, unless overridden via max_tokens in ....

• reasoning = "enabled" (extended thinking):

  • temperature must be 1. If you supply a different value in ..., this function throws an error.

  • Defaults to max_tokens = 2048 and thinking_budget_tokens = 1024, with the constraint 1024 <= thinking_budget_tokens < max_tokens. Violations of this constraint produce an error.

As a result, batches without extended thinking (reasoning = "none") use model-default sampling. When you opt into extended thinking (reasoning = "enabled"), Anthropic's requirement of temperature = 1 is enforced for all batch requests.

See also

llm_submit_pairs_batch(), llm_download_batch_results()

Other batch backends: anthropic_create_batch(), anthropic_download_batch_results(), anthropic_get_batch(), anthropic_poll_batch_until_complete(), build_gemini_batch_requests(), build_openai_batch_requests(), gemini_create_batch(), gemini_download_batch_results(), gemini_get_batch(), gemini_poll_batch_until_complete(), llm_download_batch_results(), llm_resume_multi_batches(), llm_submit_pairs_batch(), llm_submit_pairs_multi_batch(), openai_create_batch(), openai_download_batch_errors(), openai_download_batch_output(), openai_get_batch(), openai_poll_batch_until_complete(), openai_upload_batch_file(), run_anthropic_batch_pipeline(), run_gemini_batch_pipeline(), run_openai_batch_pipeline(), write_openai_batch_file()

Examples

data("example_writing_samples", package = "pairwiseLLM")

pairs <- example_writing_samples |>
  make_pairs() |>
  sample_pairs(n_pairs = 3, seed = 123) |>
  randomize_pair_order(seed = 456)

td <- trait_description("overall_quality")
tmpl <- set_prompt_template()

# Standard batch requests without extended thinking
reqs_none <- build_anthropic_batch_requests(
  pairs             = pairs,
  model             = "claude-sonnet-4-5",
  trait_name        = td$name,
  trait_description = td$description,
  prompt_template   = tmpl,
  reasoning         = "none"
)

reqs_none
#> # A tibble: 3 × 2
#>   custom_id       params          
#>   <chr>           <list>          
#> 1 ANTH_S17_vs_S12 <named list [3]>
#> 2 ANTH_S19_vs_S15 <named list [3]>
#> 3 ANTH_S01_vs_S15 <named list [3]>

# Batch requests with extended thinking
reqs_reason <- build_anthropic_batch_requests(
  pairs             = pairs,
  model             = "claude-sonnet-4-5",
  trait_name        = td$name,
  trait_description = td$description,
  prompt_template   = tmpl,
  reasoning         = "enabled"
)

reqs_reason
#> # A tibble: 3 × 2
#>   custom_id       params          
#>   <chr>           <list>          
#> 1 ANTH_S17_vs_S12 <named list [5]>
#> 2 ANTH_S19_vs_S15 <named list [5]>
#> 3 ANTH_S01_vs_S15 <named list [5]>