Add Embeddings service to the AI integration packages
#4368
Conversation
🦋 Changeset detectedLatest commit: 9c21de3 The changes in this PR will be included in the next version bump. This PR includes changesets to release 2 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
IMax153
force-pushed
the
feat/embeddings
branch
from
a60b382 to
4d71f07
Compare
| }).pipe(Layer.provide(NodeHttpClient.layerUndici)) | ||
|
|
||
| // Create an embeddings service for the `text-embedding-3-large` model | ||
| const TextEmbeddingsLarge = OpenAiEmbeddings.layerDataLoader({ |
There was a problem hiding this comment.
will it batch if I use
(documents: Array<string>) => Effect.forEach(documents, embedding.embed, { concurrency: "unbounded" })and OpenAiEmbeddings.layer?
There was a problem hiding this comment.
You would need to add batching true to the forEach options as under the hood it's using RequestResolver.batched.
Unless you use the data loader variant, in which case it will always batch as much as it can during the specified window (up to the max batch size).
| }), | ||
| Effect.map((response) => | ||
| response.data.map(({ embedding, index }) => ({ | ||
| embeddings: embedding as Array<number>, |
There was a problem hiding this comment.
should we define EmbeddingVector data type?
There was a problem hiding this comment.
I don't see a real reason to do so unless I'm missing something?
I'm just casting from a read only array to a mutable one here.
IMax153
force-pushed
the
feat/embeddings
branch
2 times, most recently
from
bf8ee12 to
d8ef64b
Compare
IMax153
force-pushed
the
feat/embeddings
branch
from
9c21de3 to
fa1948c
Compare
Type
Description
Related