Skip to content
ENExplore Calculators

Vector Database Storage Calculator

Work out vector database storage instantly with clear inputs, formula shown and shareable results.

About Vector Database Storage Calculator

Raw vector storage is simply count times dimensions times bytes per value, so five million 768-dimensional fp32 vectors need about 15 GB before any index exists. HNSW graphs add neighbour lists worth roughly 20 to 50 percent on top, and payload metadata adds more. Switching to fp16 or int8 quantisation halves or quarters the payload and is usually the first lever to pull.

Formula

Vector store size

raw = vectors x dimensions x bytes per value; total = raw x (1 + index overhead)

Worked example

With the default values below, Vector Database Storage Calculator returns Total storage required of 18.60 GiB. Change any input above and every figure here is recalculated in your browser.

Inputs used

Number of vectorsminimum 1
5,000,000
Dimensions per vectorminimum 1
768
Bytes per valueaccepts 0.5–8
4
Index and metadata overheadaccepts 0–300 %
30 %

Results

Total storage required
18.60 GiB
Raw vector payload
14.31 GiB
Index and metadata overhead
4.29 GiB
Bytes per vector all in
3,994 B

Step by step

  1. 1Values used
  2. 2Vector store size
  3. 3Total storage required= 19,968,000,000
  4. 4Raw vector payload= 15,360,000,000
  5. 5Index and metadata overhead= 4,608,000,000
  6. 6Bytes per vector all in= 3,994 B

Assumptions and limits

  • The result is derived only from the 4 inputs listed above. Anything not among them is not part of the calculation.
  • In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
  • Figures are rounded for display to whole numbers, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.

How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.

Frequently Asked Questions

How much does quantisation save?

Moving fp32 to int8 cuts the payload to a quarter with a small recall loss; product quantisation can reach 10 to 30 times compression when recall can be traded for memory.

Does the index have to fit in RAM?

For low-latency HNSW search, effectively yes. Disk-based indexes such as DiskANN exist but add tens of milliseconds per query.

Related Calculators

Vector Database Storage Calculator | Calcrivo