Repository navigation
Add the multi2vec vectorizers - #489
Merged
Merged
Conversation
![orca-security-eu[bot]](https://avatars.githubusercontent.com/in/276244?s=60&v=4){kind=small}
There was a problem hiding this comment.
Orca Security Scan Summary
| Status | Check | Issues by priority | |
|---|---|---|---|
 Passed |
Secrets |  0  0  0  0 |
View in Orca |
g-despot
force-pushed
the
v6-multi2vec
branch
from
October 15, 2026 08:13
7dcf4bc to
633ecef
Compare
g-despot
force-pushed
the
v6-multi2vec
branch
from
October 15, 2026 08:27
633ecef to
62b25bc
Compare
g-despot
force-pushed
the
v6-multi2vec
branch
from
October 15, 2026 08:28
62b25bc to
c24e190
Compare
Passing field names and theirs weights is error-prone. Marshaling a combined field is a little more involved than simply relying on the struct's format, but hey, we want a good UX
…welvelabls, voyageai, and google modules
Collaborator
|
Did some refactoring to allow users to specify properties alongside their weights: TextFields: []modules.WeightedField{
{Property: "title", Weight: 0.2},
{Property: "lyrics", Weight: 0.3},
},
ImageFields: []modules.WeightedField{
{Property: "cover", Weight: 0.5},
},This way the user doesn't need to be careful about maintaining the order the properties were passed in when giving them weights. Notice that expected values in unit tests are not changed, so the refactor worked :) Also moved all modules to |
bevzzz
reviewed
Oct 25, 2026
tsmith023
approved these changes
Oct 26, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds all 9 multimodal vectorizers: cohere, google, jinaai, nvidia, voyageai, twelvelabs, aws, clip and bind. Same conventions as the text2vec modules; builds on #484.
Field weights are plain lists, parallel to the field lists, matching the wire format. multi2vec-cohere was tested end to end against the real Cohere API (text and image objects both vectorized).