Hello, someone would please clarify if there is a limit on serverless indexes I can create with Standard plan?
I’d be happy with “pay as you go” plan but now I see you mention a limit of 20 serverless indexes per project (on Standard plan).
At the moment I have 45 indexes within one single project (hence exceeding the limit reported), and still can add more serveless indexes… therefore I don’t understand how it works.
Hey there! Let me help clarify both the limits and suggest some best practices:
First, regarding limits: On the Standard plan, you can have up to 20 serverless indexes per project according to our pricing page, but you can have up to 10,000 namespaces
In most cases where you’re considering adding multiple indexes, using namespaces is actually recommended instead. Here’s why:
Better Resource Management: Namespaces allow you to partition records within an index. All queries and operations are confined to a single namespace, providing good isolation
Simplified Management: Namespaces provide an easier way to segment your data. * Most use cases that seem to require distinct indexes can actually be handled through namespaces
For example, if you’re building a multi-tenant solution:
python
from pinecone.grpc import PineconeGRPC as Pinecone
pc = Pinecone(api_key="YOUR_API_KEY")index = pc.Index("multitenant-app")
# Upsert data for different tenants in different namespacesindex.upsert( vectors=[ {"id": "A", "values": [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1]}, {"id": "B", "values": [0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2]} ], namespace="tenant1")
Consider consolidating your indexes into fewer indexes with multiple namespaces unless you have a specific technical requirement for separate indexes. This approach will likely be more cost-effective and easier to manage in the long run.
Now that I swapped to multitenant with single serverless index and multiple namespaces… it reports you can have unlimited serveless indexes in “standard plan”.
Did Pinecone changed something?
thnks
For Standard plan it’s: “Unlimited Serverless, Inference, and Assistant usage”. Usage not numbers.
Anyway there is no tech limit to 20 as I had 60 running just 3 days ago.