FIRST Look at Pinecone Serverless!

Here we take a first look at the new Pinecone serverless, a complete rebuild of the Pinecone vector database optimized for real-world requirements, focusing on cost, latency, and recall performance.

Here, we take a look at serverless and how to use it for RAG, semantic search, or other AI applications via the Pinecone Python client.

📌 Code:

🌲 Subscribe for Latest Articles and Videos:

👾 Discord:

[carterjames]

Just finished great video again James, crazy thing is I was just reading an email about serverless from pinecone and thought I’ll just wait for James to make a video explaining this and look what we got here lol

[energyexecs]

James - Great presentation. Is the Pinecone solution "enterprise grade"? The reason for asking is I work at a large traditional utility. I am on the pilot team for AI. We don't officially allow employees to use AI. However, I submitted a Use Case to RAG and tether or augment our LLMS using RAG. The question is my company standardized on Microsoft products and I submitted my Use Case leveraging Azure (Azure Open AI API) that has lots of steps. Based on this presentation, it seems like Pinecone is way ahead of the ball game; and the path forward.

[yschermer]

Nice, what are the cold start latencies of serverless vs. pods?

[jpsl]

why using more namespaces reduce the cost?

[achukisaini]

Hi james, I am working on Bot using RAG with llama2 and for embeddings sentence transformers but I have completed the task but how to save a context or conversation in this like chatgpt coz it is a conversational bot how to do that can u please me this or tell which video I have to see?

[prashank]

We have close to 100 million vectors stored over many pods, so we’re super interested in migrating. When do you think AWS us-east option will be available and is serverless highly available?

[carterjames]

For embedding queries for similarity comparison does that factor in for how many writes you have Ina month ? **im asking before watching**

[ipranay]

How does this compare to supabase / Postgres?

[bastabey]

I guess it is the OpenAI entering the RAG space with its Retrieval Assistant disrupted the existing Vector DB players specially Pinecone.. OpenAI multi billion dollar LLM investments cannot be dependent on VecDB/RAG providers.. the old Pinecone prices were more expensive than OpenAI LLM prices which is ridiculous.. even with new prices, I bet the profit margins are very high for Pinecone

Thanks for the video

[jeffsteyn]

Cheap is awesome. But when are you going to have a complete solutions. Why do i have to go find a way to produce sparse vectors for example or even dense vectors for that matter. Imagine Microsoft selling sql server and i must find my own way to search the db or back the db. Come on guys.