2
Support higher dimension vectors
Source: supabase/supabase#29193 · opened by @julianlumets
Hi team!
I wanted to create a RAG app using embeddings with a dimension size of 3072. However i followed this tutorial ( and at the step:
create index on document_sections using hnsw (embedding vector_ip_ops);
Supabase SQL Editor returns the error:
"not more than 2000 dimensions allowed"
Could you increase the dimension size to support newer embedding models?
Thank you!
I wanted to create a RAG app using embeddings with a dimension size of 3072. However i followed this tutorial ( and at the step:
create index on document_sections using hnsw (embedding vector_ip_ops);
Supabase SQL Editor returns the error:
"not more than 2000 dimensions allowed"
Could you increase the dimension size to support newer embedding models?
Thank you!
No pledges yet. Be the first to back this.
Comments
Similar requests
Get image width/height
2 votes · 0 comments
User enumeration that could be avoided on email change
4 votes · 0 comments
SupabaseVectorStore in Langchain should update documents instead of duplicating them
1 vote · 0 comments
upgrade pgvector to 0.7
3 votes · 0 comments
Standalone Spinner Loader and Skeleton Loaders
1 vote · 0 comments
No comments yet.