Repository navigation
docs: multi-vector embeddings on the HFresh index (v1.40) - #564
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
added a commit
that referenced
this pull request
Oct 7, 2026
* docs: v1.40 version bumps — release notes, compose pins, versions-config, release history * docs: Meta AI generative model provider integration (#552) * docs: Meta AI generative model provider integration * docs: align the Meta provider page with the model-provider template * docs: mark 4-bit RQ generally available as of v1.40 (#549) * docs: named vectors by default — vector-less collections and the auto-schema default vector (#550) * docs: named vectors by default — vector-less collections and the auto-schema default vector * docs: re-ground named-vector defaults at stable/v1.40 — DEFAULT_VECTORIZER_MODULE deprecated, mixed collections, default-alias scope * docs: multi-vector embeddings on the HFresh index (v1.40) (#564) * docs: v1.40 release review fixes Correct the named-vector defaults and the DEFAULT_VECTOR_DISTANCE_METRIC scope, scope the multi-vector compression notes to the index type, fix the HFresh multi-vector reference, and refresh the patch versions and the release history C# column. Document PathSeer, the MCP hybrid changes, the backup-gcs transport default, weaviate_shards, the shards API per_node_status, read_backups, operator_metadata_reader, RQ centering, and the text2vec-google mutable settings. Add near-vector to the GraphQL migration guide and remove the temporary /deploy/enterprise redirect. * ci(links): skip doi.org and dl.acm.org, which 403 automated requests * docs: shorten the PathSeer concepts section and skip linkedin.com in link validation Align the PathSeer section with the measured results: it wins or ties on random, topical and combined filters, and ACORN stays ahead on sparse negatively correlated filters. Drop the per-index details that belong in the reference. LinkedIn rate-limits the CI link validator with 429 on the footer link. * docs: clarify the single vector collection caution * docs: drop the redundant vector-less collection how-to section * chore(specs): refresh the Weaviate OpenAPI spec to v1.40.0 Built from the v1.40.0 tag with the docs overlay. No paths or definitions were added or removed since 1.39.8. Changes land inside existing ones: read_backups in the permission enum, per_node_status on the shard status response with status and vectorQueueSize deprecated, dedupeReplicas and dedupeConvergenceTimeoutSeconds on the backup config, users and roles in the backup descriptions, license-key 403s on the namespaces endpoints, and the vectorizer empty-default wording. The overlay picks up the new collections link and the license-key sentence on the namespaces tag. * docs: remove restated v1.40 facts, keep each in its canonical home Deduplicate the DEFAULT_VECTORIZER_MODULE deprecation to the env-vars row and the migration section, shrink the vector-less collection reference, state the HFresh MUVERA requirement and search budgets once, shorten the 4-bit RQ, shards API, MCP hybrid and Meta provider additions, and correct the REST search request-body cap, which has returned 413 since v1.39.7.
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
1 participant
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.
Documents multi-vector (MUVERA) support on the HFresh index: a reference section in the vector-index config-ref (
multivector.muvera.enabledrequired and set at creation, creation-time upper limits, the has-no-effect caution formultivector.enabledalone), an hfresh-only concepts section (routing/rescore budgets, storage cost), a pointer admonition on the MUVERA compression page, and the availability clause in the vector-config how-to. Shared parameter definitions stay on their existing MUVERA page — only hfresh-specific facts are added.Grounded in a file:line CodeBrief at v1.40.0-rc.1; rendered pages verified on a local dev server.
🤖 Generated with Claude Code
https://claude.ai/code/session_01Tgpjagg1cr76MgSBS2kqrx