OpenAI embedding readiness brief
Read-only OpenAI embedding workflow. It lists visible models, creates embeddings for explicit caller-provided text samples, and summarizes vector and token metadata for semantic search or RAG readiness. It never uploads files, mutates vector stores, creates responses, enables tools, streams output, or calls arbitrary OpenAI routes.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (11 steps)
models_raw
?
embedding
?
models
default
Apply default
model_rows
slice
Take a subset of
embedding_rows
default
Apply default
embedding_count
count
Count items in embedding rows
model_count
count
Count items in models
summarize
?
model_table
to_table
Format results as a data table
embedding_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
=
Summarize whether the selected OpenAI embedding setup is ready for semantic search or RAG. Cover model visibility, embedding count, dimensions or format implied by the response, token usage, and practical follow-ups.
inputs
setting
=
openai
setting
=
dimensions
setting
=
1536
max length
setting
=
900
embedding model
setting
=
text-embedding-3-small
encoding format
setting
=
float
model row limit
setting
=
20
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.