Package index • LLMR

Core

llm_config(): Create an LLM configuration (provider-agnostic)
print(<llm_config>) format(<llm_config>): Print an LLM configuration with the API key masked
llm_api_key_env(): Declare an API key sourced from an environment variable
call_llm(): Call an LLM (chat/completions or embeddings) with optional multimodal input
call_llm_stream(): Stream a chat completion token by token
finish_reason() tokens() is_truncated() as.character(<llmr_response>) print(<llmr_response>): LLMR Response Object
chat_session() as.data.frame(<llm_chat_session>) summary(<llm_chat_session>) head(<llm_chat_session>) tail(<llm_chat_session>) print(<llm_chat_session>): Chat Session Object and Methods

Structured output

enable_structured_output(): Enable Structured Output (Provider-Agnostic)
disable_structured_output(): Disable Structured Output (clean provider toggles)
llm_parse_structured(): Parse structured output emitted by an LLM
llm_parse_structured_col(): Parse structured fields from a column into typed vectors
llm_parse_tags(): Parse XML-like tags emitted by an LLM
llm_parse_tags_col(): Parse XML-like tag fields from a column
llm_parse_rowpack_tags(): Parse a batched, row-wrapped tag response into per-row field lists
llm_validate_structured_col(): Validate structured JSON objects against a JSON Schema (locally)
llm_fn_structured(): Vectorized structured-output LLM
llm_mutate_structured(): Data-frame mutate with structured output
llm_mutate_tags(): Data-frame mutate with XML-like tag output
llm_fn_tags(): Vectorized LLM with tag extraction
call_llm_par_structured(): Parallel experiments with structured parsing
call_llm_par_tags(): Parallel experiments with tag parsing

Inspecting and summarizing runs

llm_preview(): Preview a tidy LLM call without spending anything
llm_render_messages(): Render tidy messages without calling any API
transcript_as_messages(): Build a role-flipped message array from a multi-speaker transcript
ensure_alternating_messages(): Make a message array provider-safe (alternating, user-leading)
llm_usage(): Summarize token usage and outcomes of an LLM run
llm_failures(): List the rows of an LLM run that failed or were truncated
llm_methods_text(): Draft a methods-section paragraph from an LLM run
llm_log_enable() llm_log_disable() llm_log_status() llm_log_active() llm_log_merge(): Record every LLM call in a local audit log
llm_log_read(): Read an LLMR audit log into records and a manifest
llm_logprobs(): Extract token log-probabilities from a response
llm_hash(): Content hash for research artifacts
llm_uuid(): A short, sortable, process-unique identifier

Provenance and replay

Turning an audit log back into a reproducible request: a content hash of a request, the response record it produced, and the reconstruction of a request from a logged call.

llm_request_hash(): Stable request hash for an LLM call
llm_request_from_log(): Rebuild a callable request from a logged record
llm_response_record(): Flatten one LLM response to a provenance row

Method-package generics

Shared generics implemented by the LLMR method packages (LLMRcontent, LLMRpanel, FocusGroup) on their result objects.

diagnostics(): Machine-readable diagnostics for an LLMR-family result object
report(): Draft a methods-section report from an LLMR-family result object
reset(): Reset a stateful object to its initial position

Tidy helpers and parallel

llm_fn(): Apply an LLM prompt over vectors/data frames
llm_mutate(): Mutate a data frame with LLM output
llm_replicate(): Run the same prompt several times per row
llm_agreement(): Agreement across replicated LLM annotations
build_factorial_experiments(): Build Factorial Experiment Design
call_llm_par(): Parallel LLM Processing with Tibble-Based Experiments (Core Engine)
llm_add_request_hash(): Append the audit-log request hash to a parallel-results frame
call_llm_broadcast(): Parallel API calls: Fixed Config, Multiple Messages
call_llm_sweep(): Parallel API calls: Parameter Sweep - Vary One Parameter, Fixed Message
call_llm_compare(): Parallel API calls: Multiple Configs, Fixed Message
expand_llm_config(): Expand an LLM Config Grid
llm_cross_design(): Cross a data frame with LLM configs
llm_par_resume(): Resume failed parallel LLM calls
llm_judge(): LLM-as-a-Judge Evaluation
setup_llm_parallel(): Setup Parallel Environment for LLM Processing
reset_llm_parallel(): Reset Parallel Environment

Batch APIs (asynchronous, discounted)

llm_batch_submit(): Submit a batch job to a provider's batch API
llm_batch_status(): Check the status of a batch job
llm_batch_fetch(): Fetch the results of a batch job
llm_batch_cancel(): Cancel a batch job

Tools (function calling)

llm_tool(): Define a tool the model may call
llm_tool_signature(): A stable signature for a tool's identity
call_llm_tools(): Call an LLM with tools and run the tool loop
tool_calls(): Extract tool calls from a response
bind_tools(): Bind tools to a config (provider-agnostic)

Embeddings

parse_embeddings(): Parse Embedding Response into a Numeric Matrix
get_batched_embeddings(): Generate Embeddings in Batches

Reliability

call_llm_robust(): Robustly Call LLM API (Simple Retry)
cache_llm_call(): Cache LLM API Calls
log_llm_error(): Log LLMR Errors

Persona data

A bundled example persona dataset and the small contract helpers that read such frames, shared across the LLMR family.

anes_2024_personas: Example participant profiles derived from ANES 2024
llm_persona_split(): Split one persona into labeled demographics and labeled answers
llm_persona_overview(): A compact overview of a persona frame for display
llm_persona_dictionary(): The question dictionary attached to a persona data frame
llm_persona_demographic_fields(): Which columns of a persona frame are demographics
llm_validate_persona_frame(): Check that a data frame follows the persona contract