pg_vectorize
pg_vectorize copied to clipboard
tembo-io
The simplest way to build AI workloads on Postgres
add support for embeddings from: - aws bedrock - google vertex - [azure ai](https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/embeddings?tabs=console#how-to-get-embeddings) - [voyage](https://docs.voyageai.com/docs/embeddings) Embedding providers are added by implementing the required traits. For example see the implementation...
Provide ability to automatically chunk text in the input columns to the vectorize.table function, or provide a utility function (`vectorize.chunk_table()`?) that takes an input table, chunks the data in each...
`vectorize.search()` supports some limited filtering (the filtering happens after vector similarity search) by using the [where](https://github.com/tembo-io/pg_vectorize/blob/9e013e3c9c5702c982ca74415ff396074a744cb7/extension/src/api.rs#L54) parameter, but we dont have any docs for this. 1. Add usage of this...
Make pg_vectorize work with postgres 17. There will be a handful of steps to make this possible. 1. update [pgrx to >= v0.12.5 ](https://github.com/tembo-io/pg_vectorize/blob/9e013e3c9c5702c982ca74415ff396074a744cb7/extension/Cargo.toml#L24) 2. add pg17 feature to [Cargo.toml](https://github.com/tembo-io/pg_vectorize/blob/9e013e3c9c5702c982ca74415ff396074a744cb7/extension/Cargo.toml#L14)...
```sql SELECT vectorize.table( job_name => 'product_search_hf', "table" => 'products', primary_key => 'product_id', columns => ARRAY['product_name', 'description'], transformer => 'sentence-transformers/all-MiniLM-L6-v2', schedule => 'realtime' ); ``` Having "table" as parameter causes a...
`search_alg` is deprecated, in favor of [index_dist_type](https://github.com/tembo-io/pg_vectorize/blob/9e013e3c9c5702c982ca74415ff396074a744cb7/extension/src/api.rs#L22). Drop search_alg from the parameter and downstream call tree.
if a table is dropped that is part of vectorize.jobs, then the vectorize.job should also be deleted. an event trigger can solve this. some scratch code as example of the...
provide a feature or tooling to allow a user to take embeddings from one table and make it such that pg_vectorize can manage those embeddings. for example, assume a user...
update https://github.com/tembo-io/pg_vectorize/blob/9e013e3c9c5702c982ca74415ff396074a744cb7/vector-serve/pyproject.toml#L12 to the latest version. Make sure `poetry.lock` gets updated as well. Probably good idea to update the rest of the python dependencies as well. Merging will build and...