Limits

There isn't a limit or performance metric we can't improve by an order of magnitude when prioritized. If you expect to brush up against a limit or you are limited by present performance, contact us.

LimitCurrent
Max documents (global)
Unlimited (seen: 4T+ @ 15PB+)
Max documents (per namespace)
128B @ 256TB (seen: 100B @ 200TB)
Max shards per namespace
256
Max documents (per shard)
500M @ 1TB
Max number of namespaces
Unlimited (seen: 250M+)
Max number of pinned namespaces
256
Max vector columns per namespace
2
4
Max dimensions for dense vectors
10,752
Max total dimensions for sparse vectors
Unlimited (seen: 30,522)
Max dimensions per sparse vector
1,024
Max inactive time in cache
hours
Max write throughput (global)
Unlimited (seen: 10M+ writes/s @ 32GB/s)
Max write throughput (per namespace)
10k writes/s @ 32 MB/s
Max namespace copy throughput
Contact us if bottlenecked
Number of branches
Unlimited (seen: 10M+)
Max upsert batch request size
512 MB
Max upsert batch of embedded attributes
30 rows
Max rows affected by patch by filter
50k
Max rows affected by delete by filter
5M
Max ingested, unindexed data
2 GB
Max queries (global)
Unlimited (seen: 25k+ queries/s)
Max queries (per namespace)
5k+ queries/s
Max queries in a multi-query request
16
Max concurrent queries per namespace
16 (100s of queries/s)
Max read replicas
Unlimited (seen: 3)
Vector search recall@10
90-100%
Max attribute value size
8 MiB
Max filterable value size
4 KiB
Max document size
64 MiB
Max id size
64 bytes
Max attribute name length
128 bytes
Max attribute names per namespace
1,024
Max namespace name length
128 bytes
Max full-text query length
1,024
10k
Max aggregation groups per query
10k
Max computed attributes per query
256
Max fragments per highlight
1024