m-backend-submit.md
In mall: Run Multiple Large Language Model Predictions Against a Table, or Vectors

Ollama code is covered

Code
  llm_vec_sentiment(test_txt)
Message
  ! 15 records may be over 10 tokens
  Ollama may have truncated what was sent to the model 
  (https://github.com/ollama/ollama/issues/7043)
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  * positive...
  i 5 more records
Output
   [1] "positive" "positive" "positive" "positive" "positive" "positive"
   [7] "positive" "positive" "positive" "positive" "positive" "positive"
  [13] "positive" "positive" "positive"

ellmer code is covered with cache turned off

Code
  m_backend_submit(backend = ellmer_session, x = test_txt, prompt = list(list(
    content = "test")), preview = TRUE)
Output
  [[1]]
  ellmer_obj$set_system_prompt(list(list(content = "test")))

  [[2]]
  ellmer_obj$chat(as.list("test"))