Hugging Face Hub discovery brief
Use “Hugging Face Hub discovery brief” in Recued. It provides a read-only Hugging Face Hub discovery workflow.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (22 steps)
normalized_query
trim
Trim whitespace from setting query
quicksearch
?
trending
?
docs
?
docs_full_text
?
hardware
?
defaults
defaults
Apply defaults
model_count
count
Count items in models
dataset_count
count
Count items in datasets
space_count
count
Count items in spaces
paper_count
count
Count items in papers
collection_count
count
Count items in collections
summarize
?
model_rows
slice
Take a subset of
space_rows
slice
Take a subset of
paper_rows
slice
Take a subset of
doc_rows
slice
Take a subset of
model_table
to_table
Format results as a data table
space_table
to_table
Format results as a data table
paper_table
to_table
Format results as a data table
docs_table
to_table
Format results as a data table
card
to_summary
Format results as a summary card
Settings
Configurable at install. Defaults shown — change them anytime in Recued.
focus
setting
=
Identify the most useful Hugging Face resources for this topic. Compare models, datasets, Spaces, papers, collections, relevant documentation, and practical next steps. Do not invent facts outside the Hugging Face results.
query
setting
=
row limit
setting
=
10
max length
setting
=
900
trend type
setting
=
models
doc product
setting
=
hub
huggingface
setting
=
result limit
setting
=
10
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.