feat(helm): add persistent volume for local model cache#861
Merged
nicoloboschi merged 2 commits intovectorize-io:mainfrom Apr 7, 2026
Merged
feat(helm): add persistent volume for local model cache#861nicoloboschi merged 2 commits intovectorize-io:mainfrom
nicoloboschi merged 2 commits intovectorize-io:mainfrom
Conversation
When using local reranker (e.g., BAAI/bge-reranker-v2-m3) or local embedding models, the models are downloaded to /home/hindsight/.cache on every pod restart, causing slow startup and unnecessary bandwidth. Add optional persistent volume support: - api: PVC mounted at /home/hindsight/.cache - worker: volumeClaimTemplate (StatefulSet) at same path Disabled by default. Enable via: api.persistence.modelCache.enabled: true worker.persistence.modelCache.enabled: true Closes vectorize-io#860 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
isac322
force-pushed
the
feat/helm-model-cache-volume
branch
from
ab2e152 to
4de08f4
Compare
Allow users to mount arbitrary volumes (configMaps, secrets, emptyDir, etc.) into api and worker pods via values, following common helm chart library conventions. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
nicoloboschi
approved these changes
Apr 7, 2026
Collaborator
nicoloboschi
left a comment
nicoloboschi
left a comment
There was a problem hiding this comment.
awesome, thank you!
I wanted to add this some time ago but couldn't find the time, thanks!!!
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
When using local reranker models (e.g.,
BAAI/bge-reranker-v2-m3) or local embedding models, the models are downloaded to/home/hindsight/.cacheon every pod restart. This causes slow startup (~1GB+ download) and unnecessary bandwidth usage.Closes #860
Changes
api.persistence.modelCacheandworker.persistence.modelCacheconfig (disabled by default)/home/hindsight/.cachewhen enabledvolumeClaimTemplatesfor model cache when enabledUsage
Notes
volumeClaimTemplates(StatefulSet) for per-replica storage