Skip to contents

Check whether a local Ollama server is reachable

Source: R/llm_kg_pipeline.R
llm_kg_ollama_check.Rd

Issues a 1-second HEAD request to the Ollama /api/tags endpoint and reports whether the server responds. Used by llm_kg_pipeline_run() as a pre-flight gate.

Usage

llm_kg_ollama_check(
  host = "http://localhost:11434",
  model = NULL,
  timeout_sec = 1
)

Arguments

host

Character; the base URL of the Ollama server. Default "http://localhost:11434".

model

Optional character; if supplied, additionally verifies that the named model is present on the server.

timeout_sec

Numeric; HTTP timeout. Default 1.

Value

Named logical list with reachable (bool) and (when model is non-NULL) model_present (bool).