Skip to content

Benchmarks

Performance results for Calagopus, measured against five other panels on identical hardware. Each chart compares every panel across every test configuration.

Click a panel to show or hide it in the charts below.

peak throughput · 8c
245k req/s
245kCalagopus
177kPufferPanel
1,431FeatherPanel
1,170Pterodactyl
799Hydrodactyl
381Pelican
avg response · 8c
1.6 ms
1.6 msCalagopus
2.2 msPufferPanel
270 msFeatherPanel
333 msPterodactyl
491 msHydrodactyl
1066 msPelican
peak memory · 8c
92 MiB
77 MiBPufferPanel
92 MiBCalagopus
148 MiBPterodactyl
158 MiBHydrodactyl
261 MiBFeatherPanel
358 MiBPelican

Methodology

All panels run from their official :latest Docker images (as of 2026-07-21) with no configuration beyond initial setup. Runs are driven by our open-source benchmarking suite: each panel boots from its own Docker Compose stack and is swept across CPU-quota variants (1, 2, 4, and 8 CPUs) while container CPU and memory are sampled. Every scenario runs 384 concurrent connections for 10 seconds after a 1-second warmup, against three endpoints: public settings (unauthenticated), account details (authenticated), and the server list (authenticated). All panels keep their default rate limiting; high [429] counts on authenticated endpoints are expected.

Configurations are labeled <CPU> · <cores>c. The numbers on this page are the suite's JSON reports (pnpm run bench <panel> --json) checked into the site as typed data - anyone can reproduce a run and compare.

Test environments

Ryzen 9 9900XAMD Ryzen 9 9900X · DDR5-6000
panelpeak cpupeak mem
Calagopus812%84 MiB
FeatherPanel824%237 MiB
Hydrodactyl819%235 MiB
Pelican525%389 MiB
Pterodactyl818%212 MiB
PufferPanel548%119 MiB
EPYC 7443PAMD EPYC 7443P · DDR4-2666
panelpeak cpupeak mem
Calagopus758%138 MiB
FeatherPanel827%266 MiB
Hydrodactyl823%212 MiB
Pelican591%433 MiB
Pterodactyl816%194 MiB
PufferPanel588%120 MiB
2× Xeon E5-2680 v22× Intel Xeon E5-2680 v2 · DDR3-1600
panelpeak cpupeak mem
Calagopus754%92 MiB
FeatherPanel826%229 MiB
Hydrodactyl817%237 MiB
Pelican553%441 MiB
Pterodactyl820%209 MiB
PufferPanel530%85 MiB
Core i5-12500Intel Core i5-12500 · DDR4-3200
panelpeak cpupeak mem
Calagopus550%137 MiB
FeatherPanel827%262 MiB
Hydrodactyl818%194 MiB
Pelican528%413 MiB
Pterodactyl819%238 MiB
PufferPanel472%116 MiB
Ampere Altra Q80-30Ampere Altra Q80-30 · DDR4-2133
panelpeak cpupeak mem
Calagopus367%80 MiB
FeatherPanel837%261 MiB
Hydrodactyl812%223 MiB
Pelican543%434 MiB
Pterodactyl813%206 MiB
PufferPanel570%77 MiB

Peaks are the highest values observed across all CPU configs and scenarios on that system.

Memory usage

Mean under load

↓ Lower is better - less memory used while serving load.

Peak under load

↓ Lower is better - smaller worst-case memory footprint.

Throughput

Scenario: settings (unauth)

↑ Higher is better - more requests served per second.

Scenario: account (auth)

↑ Higher is better - more requests served per second.

Scenario: list servers (auth)

↑ Higher is better - more requests served per second.

Average latency

Scenario: settings (unauth)

↓ Lower is better - faster average response.

Scenario: account (auth)

↓ Lower is better - faster average response.

Scenario: list servers (auth)

↓ Lower is better - faster average response.

p99 latency

Scenario: settings (unauth)

↓ Lower is better - faster worst-case (p99) response.

Scenario: account (auth)

↓ Lower is better - faster worst-case (p99) response.

Scenario: list servers (auth)

↓ Lower is better - faster worst-case (p99) response.

Detailed comparisons

Raw numbers are only part of the picture. For feature-by-feature breakdowns against specific panels, see Calagopus vs Pterodactyl, Calagopus vs Pelican, and Calagopus vs AMP.