Real-world vector DB performance across the most popular providers

[yggdrasill501]

Best vector DB benchmark I have seen, solid benchmark design, but would be good if you would have shown who are the competitors in the graphs instead of anonymizing the numbers.

[jerguslejko]

Hi, I'm Jergus, one of the founders of TopK. We cannot share the results publicly but happy to share privately (@jerguslejko on twitter, or [e-mail])

[MarekDlugos]

My friends benchmarked managed vector databases under production-like conditions: high-throughput ingest, concurrent queries, filtering, and read–write mixed workloads.

The post includes the methodology, the dataset, and the open-source tool they published for running the benchmarks.

[TechIsCool]

Feels like a sales pitch only due to the abstraction of Provider A,B,C vs actually naming the products. Guess thats what you get for a vendor blog.

[jerguslejko]

Hey, author of the post here.

We're actually not allowed to post head to head comparison with competitors and share their names, that's why :) Post contains the dataset, the tool and methodology how the data was collected, which hopefully gives confidence in fairness of the benchmark.

[ms1472]

Is pgvector one of the systems you tested, or was it intentionally left out?

[jerguslejko]

We didn’t include pgvector because we focused on managed services to keep things comparable — TopK is managed/serverless, so the fair match would be a managed Postgres. And pgvector just doesn’t really scale to the kinds of workloads we ran here.