llm_config: Create an LLM configuration (provider-agnostic)

In LLMR: Interface for Large Language Model APIs in R

Create an LLM configuration (provider-agnostic)

Description

llm_config() builds a provider-agnostic configuration object that call_llm() (and friends) understand. You can pass provider-specific parameters via ...; LLMR forwards them as-is, with a few safe conveniences.

Usage

llm_config(
  provider,
  model,
  api_key = NULL,
  troubleshooting = FALSE,
  base_url = NULL,
  embedding = NULL,
  no_change = FALSE,
  ...
)

Arguments

provider	Character scalar. One of: "openai", "anthropic", "gemini", "groq", "together", "voyage" (embeddings only), "deepseek", "xai".
model	Character scalar. Model name understood by the chosen provider. (e.g., "gpt-4o-mini", "o4-mini", "claude-3.7", "gemini-2.0-flash", etc.)
api_key	Character scalar. Provider API key.
troubleshooting	Logical. If TRUE, prints the full request payloads (including your API key!) for debugging. Use with extreme caution.
base_url	Optional character. Back-compat alias; if supplied it is stored as api_url in model_params and overrides the default endpoint.
embedding	NULL (default), TRUE, or FALSE. If TRUE, the call is routed to the provider's embeddings API; if FALSE, to the chat API. If NULL, LLMR infers embeddings when model contains "embedding".
no_change	Logical. If TRUE, LLMR never auto-renames/adjusts provider parameters. If FALSE (default), well-known compatibility shims may apply (e.g., renaming OpenAI's max_tokens → max_completion_tokens after a server hint; see call_llm() notes).
...	Additional provider-specific parameters (e.g., temperature, top_p, max_tokens, top_k, repetition_penalty, reasoning_effort, api_url, etc.). Values are forwarded verbatim unless documented shims apply.

Value

An object of class c("llm_config", provider). Fields: provider, model, api_key, troubleshooting, embedding, no_change, and model_params (a named list of extras).

Temperature range clamping

Anthropic temperatures must be in ⁠[0, 1]⁠; others in ⁠[0, 2]⁠. Out-of-range values are clamped with a warning.

Endpoint overrides

You can pass api_url (or ⁠base_url=⁠ alias) in ... to point to gateways or compatible proxies.

See Also

call_llm, call_llm_robust, llm_chat_session, call_llm_par, get_batched_embeddings

Examples

## Not run: 
# Basic OpenAI config
cfg <- llm_config("openai", "gpt-4o-mini",
temperature = 0.7, max_tokens = 300)

# Generative call returns an llmr_response object
r <- call_llm(cfg, "Say hello in Greek.")
print(r)
as.character(r)

# Embeddings (inferred from the model name)
e_cfg <- llm_config("gemini", "text-embedding-004")

# Force embeddings even if model name does not contain "embedding"
e_cfg2 <- llm_config("voyage", "voyage-large-2", embedding = TRUE)

## End(Not run)

LLMR documentation built on Aug. 26, 2025, 9:08 a.m.