toolchest_

Measured on a real homelab

Kavita RAM & CPU requirements (measured)

On our homelab it uses about 48 MB of RAM on a normal hour, 96 MB in busy hours, and peaked at 485 MB over the last 30 days (small library: 35 books and comics (1.1 GB)).

What it really uses (process memory)

Typical (median)
48 MB
Average
57 MB
Busy hours (95th percentile)
96 MB
Highest recorded
485 MB
As docker stats shows it (with disk cache)
median 133 MB · peak 667 MB
Window
720 hours measured, up to Sep 26, 2026
Version
0.9.1.4
Stack (all summed)
kavita
Workload
small library: 35 books and comics (1.1 GB)
Hardware
Intel i5-10600K · 23 GB RAM · Docker (Coolify)

Caveat (Oct 2, 2026): these figures count process memory held in RAM. When the server is under memory pressure, idle apps get part of their memory swapped out and it was not being counted, so the "typical" figure may be too low; the peak is reliable. We now measure RAM plus swap and will update these numbers.

What the project says

Minimum RAM
256 MB
Recommended RAM
not published
CPU cores
1

Not a formal requirement: users report running it on a single 1 GHz core with 256 MB. The CPU must support SSE4.2.

Sources

What makes it grow

Memory grows with library size and with scans of new content. The official 256 MB is a user report, not a requirement: even with a small library its processes peaked at 485 MB (667 MB counting disk cache), so the resource calculator recommends 1 GB.

Sizing it

Our rule of thumb: plan for the highest process memory recorded plus 25% headroom: 640 MB. Our workload is small — scale it with the factors above before buying hardware.

Combine it with other apps in the resource calculator →

Build it: Rent a VPS · Buy a mini-PC

How we measure

Memory per container from Netdata on the same server, summed across the whole stack (app, database and cache) hour by hour. The main figures are the processes' own memory (anonymous memory: heap, stacks). The docker stats figure is higher because it also counts active disk cache — which Linux gives back when it needs it — and shared memory such as PostgreSQL's shared buffers. Hours where the main container had no data are left out, not counted as zero. Re-measured with a script, not by hand.

Other measured apps