Skip to content

BENCHMARKS.md

Latest commit

 

History

History
205 lines (152 loc) · 6.83 KB

BENCHMARKS.md

File metadata and controls

205 lines (152 loc) · 6.83 KB

Performance Benchmarks

Comprehensive benchmarks comparing ts-mocker against other popular JavaScript/TypeScript faker libraries.

Latest Results

Environment:

  • Runtime: Bun v1.2.21
  • Platform: macOS ARM64 (Apple Silicon)
  • Date: 2025-10-04

Summary

✅ ts-mocker won 9 out of 9 benchmarks ⚡ 6.57x faster than @faker-js/faker on average

Detailed Results

📛 Full Name Generation

Library ops/s avg time margin vs fastest
ts-mocker 31.35M ops/s 0.0319ms ±0.47% (fastest) ✅
@ngneat/falso 3.76M ops/s 0.2659ms ±0.95% 0.12x
casual 2.61M ops/s 0.3833ms ±0.96% 0.08x
chance 2.56M ops/s 0.3900ms ±1.01% 0.08x
@faker-js/faker 537.95K ops/s 1.8589ms ±1.40% 0.02x

📧 Email Generation

Library ops/s avg time margin vs fastest
ts-mocker 1.96M ops/s 0.5098ms ±1.43% (fastest) ✅
@ngneat/falso 1.81M ops/s 0.5512ms ±1.12% 0.92x
casual 1.09M ops/s 0.9172ms ±0.79% 0.56x
chance 849.49K ops/s 1.1772ms ±1.43% 0.43x
@faker-js/faker 567.44K ops/s 1.7623ms ±1.16% 0.29x

📞 Phone Number Generation

Library ops/s avg time margin vs fastest
ts-mocker 9.77M ops/s 0.1024ms ±1.13% (fastest) ✅
chance 2.68M ops/s 0.3735ms ±1.21% 0.27x
@faker-js/faker 2.05M ops/s 0.4869ms ±0.99% 0.21x
casual 1.23M ops/s 0.8117ms ±0.61% 0.13x
@ngneat/falso 470.39K ops/s 2.1259ms ±0.87% 0.05x

🏙️ City Generation

Library ops/s avg time margin vs fastest
ts-mocker 39.63M ops/s 0.0252ms ±0.58% (fastest) ✅
@ngneat/falso 18.13M ops/s 0.0552ms ±1.20% 0.46x
casual 1.99M ops/s 0.5020ms ±0.89% 0.05x
chance 1.12M ops/s 0.8969ms ±1.19% 0.03x
@faker-js/faker 779.69K ops/s 1.2826ms ±1.29% 0.02x

🌍 Country Generation

Library ops/s avg time margin vs fastest
ts-mocker 38.55M ops/s 0.0259ms ±0.71% (fastest) ✅
casual 18.98M ops/s 0.0527ms ±1.08% 0.49x
@ngneat/falso 17.93M ops/s 0.0558ms ±1.14% 0.46x
@faker-js/faker 15.83M ops/s 0.0632ms ±0.80% 0.41x
chance 2.53M ops/s 0.3946ms ±1.47% 0.07x

🏢 Company Name Generation

Library ops/s avg time margin vs fastest
ts-mocker 37.64M ops/s 0.0266ms ±1.15% (fastest) ✅
@ngneat/falso 17.78M ops/s 0.0562ms ±1.09% 0.47x
casual 2.32M ops/s 0.4317ms ±0.99% 0.06x
chance 800.45K ops/s 1.2493ms ±1.61% 0.02x
@faker-js/faker 557.33K ops/s 1.7943ms ±1.66% 0.01x

🛍️ Product Generation

Library ops/s avg time margin vs fastest
ts-mocker 30.65M ops/s 0.0326ms ±0.98% (fastest) ✅
casual 21.34M ops/s 0.0469ms ±0.65% 0.70x
@faker-js/faker 17.78M ops/s 0.0562ms ±0.49% 0.58x
@ngneat/falso 365.26K ops/s 2.7378ms ±0.83% 0.01x

🔑 UUID Generation

Library ops/s avg time margin vs fastest
ts-mocker 1.02M ops/s 0.9849ms ±0.26% (fastest) ✅
@ngneat/falso 555.23K ops/s 1.8011ms ±0.52% 0.55x
@faker-js/faker 531.61K ops/s 1.8811ms ±0.55% 0.52x
casual 464.73K ops/s 2.1518ms ±0.53% 0.46x
chance 428.55K ops/s 2.3335ms ±0.78% 0.42x

📊 Complex User Object (10,000 records)

Generates 10,000 complete user objects with:

  • Full Name
  • Email
  • Phone Number
  • City
  • Country
Library Total Time μs/record vs fastest
ts-mocker 7.97ms 0.7975μs (fastest) ✅
casual 25.44ms 2.5442μs 0.31x
@ngneat/falso 28.64ms 2.8644μs 0.28x
chance 33.12ms 3.3117μs 0.24x
@faker-js/faker 52.39ms 5.2394μs 0.15x

Libraries Tested

Library Version Published Size Files
ts-mocker (core) 174 KB 38
casual v1.6.2 408 KB 75
@ngneat/falso v7.4.0 1.16 MB 399
chance v1.1.13 2.08 MB 164
@faker-js/faker v10.0.0 4.29 MB 231

Package Size Advantages

ts-mocker is dramatically smaller than all competitors:

  • 24.7x smaller than @faker-js/faker (saves 4.12 MB)
  • 12.0x smaller than chance (saves 1.91 MB)
  • 6.7x smaller than @ngneat/falso (saves 1.01 MB)
  • 2.4x smaller than casual (saves 234 KB)

Even compared to the smallest competitor (casual), ts-mocker is 2.4x smaller while offering 26 full locales vs limited support.

📦 See PACKAGE-SIZES.md for detailed size comparison and analysis.

Key Takeaways

Where ts-mocker Excels

  1. Full name generation - 31.35M ops/s (8.3x faster than nearest competitor)
  2. Phone number generation - 9.77M ops/s (3.6x faster than nearest competitor)
  3. Simple data types (City, Country, Company, Product) - 30-40M ops/s
  4. Email generation - 1.96M ops/s (fastest)
  5. Complex objects - 6.57x faster than @faker-js/faker
  6. Consistency - Low margin of error (±0.47-1.43%)
  7. 100% win rate - Fastest in ALL categories

Performance vs @faker-js/faker

ts-mocker is significantly faster across all categories:

  • Full Name: 58.3x faster
  • Phone Number: 4.8x faster
  • City: 50.8x faster
  • Country: 2.4x faster
  • Company: 67.5x faster
  • Product: 1.7x faster
  • UUID: 1.9x faster
  • Email: 3.5x faster
  • Complex Objects: 6.6x faster

Average improvement: 6.57x faster

Running Benchmarks

# From repository root
bun run bench

# From benchmarks package
cd packages/benchmarks
bun run bench

Benchmark Configuration

  • Time per benchmark: 500ms
  • Warmup: Yes (automatic)
  • Iterations: Automatic (based on time budget)
  • Benchmarking library: tinybench v2.9.0

Why ts-mocker is Fast

  1. Direct array access - O(1) constant-time lookups
  2. Minimal overhead - Lightweight abstraction layer
  3. No complex computations - Simple random selection
  4. Efficient data structures - Optimized locale data
  5. Modern JavaScript - Uses latest ES features
  6. Tree-shakeable - Only load what you use

Contributing

To update benchmarks or add new tests:

  1. Edit packages/benchmarks/src/index.ts
  2. Run bun run bench to verify
  3. Update this file with new results

License

MIT