Skip to contents

Central registry of the default model used by each `huggingfaceR` inference function. Every `hf_*` function that takes a `model` argument resolves its default through this single function, so default models live in exactly one place and can be audited or updated without hunting through the codebase.

Usage

hf_default_model(task = NULL)

Arguments

task

Character string naming the task, or `NULL`. One of: "chat", "generate", "fill_mask", "classify", "zero_shot", "embed", "summarize", "translate", "ner", "question_answer", "table_question_answer", "vision_chat", "transcribe", "text_to_speech", "text_to_image", "classify_image", "caption_image", "detect_objects". When `NULL` (the default), the full registry is returned as a tibble.

Value

When `task` is supplied, a single model-ID character string. When `task` is `NULL`, a tibble with columns `task` and `model` listing every default.

Details

Defaults are chosen to be **beginner-friendly**: broadly known, small and fast, low cost, and usable with no extra arguments — the goal is the quickest path to a working first call (think of `mtcars` in base R). Power users can always override any default by passing their own `model`.

Examples

# The default model for a given task
hf_default_model("translate")
#> [1] "Helsinki-NLP/opus-mt-en-fr"

# The whole registry at a glance
hf_default_model()
#> # A tibble: 18 × 2
#>    task                  model                                                  
#>    <chr>                 <chr>                                                  
#>  1 chat                  meta-llama/Llama-3.1-8B-Instruct                       
#>  2 generate              meta-llama/Llama-3.1-8B-Instruct                       
#>  3 fill_mask             google-bert/bert-base-uncased                          
#>  4 classify              distilbert/distilbert-base-uncased-finetuned-sst-2-eng…
#>  5 zero_shot             facebook/bart-large-mnli                               
#>  6 embed                 BAAI/bge-small-en-v1.5                                 
#>  7 summarize             facebook/bart-large-cnn                                
#>  8 translate             Helsinki-NLP/opus-mt-en-fr                             
#>  9 ner                   dslim/bert-base-NER                                    
#> 10 question_answer       deepset/roberta-base-squad2                            
#> 11 table_question_answer google/tapas-base-finetuned-wtq                        
#> 12 vision_chat           google/gemma-3-4b-it                                   
#> 13 transcribe            openai/whisper-large-v3                                
#> 14 text_to_speech        facebook/mms-tts-eng                                   
#> 15 text_to_image         black-forest-labs/FLUX.1-schnell                       
#> 16 classify_image        google/vit-base-patch16-224                            
#> 17 caption_image         google/gemma-3-4b-it                                   
#> 18 detect_objects        facebook/detr-resnet-50