Performance Benchmarks

Real measurements comparing ApiArk against Postman, Bruno, Insomnia, and Hoppscotch. Installer sizes are fetched live from GitHub Releases. Internal benchmarks powered by Criterion.rs.

65 MB
RAM at idle (vs Postman 350 MB)
1.2s
Startup time (vs Postman 12s)
22 MB
Installer size (vs Postman 200 MB)
7
Protocols supported

Installer Size

Linux Installer Download Size (MB)

Fetching live data...
ApiArk: 22 MB — fallbackPostman: 200 MB — manualInsomnia: 140 MB — fallbackBruno: 85 MB — fallbackHoppscotch: 25 MB — fallback

Memory Usage

RAM at Idle (MB)

RAM with 10 Open Tabs (MB)

Startup Time

Cold Start to Ready (seconds)

Feature Comparison

FeatureApiArkPostmanInsomniaBrunoHoppscotch
FrameworkTauri v2ElectronElectronElectronTauri
Open Source
Account Required
RAM (idle)65 MB350 MB250 MB180 MB55 MB
Startup Time1.2s12s6s4.5s1.8s
Installer Size...............
REST
GraphQL
gRPC
WebSocket
SSE
MQTT
Socket.IO

Internal Benchmarks (Criterion.rs)

Measured on every commit to main. These are Rust backend benchmarks — no GUI overhead.

YAML Parse (single)
Parse one request YAML file
9.8µs
YAML Parse (complex)
Parse request with all fields
24µs
Collection Load (100)
Load 100-request collection tree
1.0ms
Collection Load (1K)
Load 1000-request collection tree
10.9ms
Interpolation (5 vars)
Replace 5 {{variables}} in URL
75µs
cURL Parse (complex)
Parse cURL with auth + headers
2.6µs
Script Engine (empty)
QuickJS runtime + ark API init
519µs
Script (4 tests)
Parse JSON + run 4 assertions
633µs
Assertions (20)
Evaluate 20 declarative assertions
38µs
Assertions (regex)
Single regex match assertion
30µs

Methodology

Installer sizes are fetched live from each tool's GitHub Releases API (latest stable release, Linux AppImage). Postman is closed-source, so its size is measured manually from the official download.

Internal benchmarks are run via Criterion.rs in CI on every push to main. Results are deterministic and reproducible.

RAM and startup times are measured on comparable hardware (8-core, 16 GB RAM, SSD) from community benchmarks and official documentation. These cannot be automated via GitHub API and are updated periodically.

If you find inaccuracies, please open an issue.