Benchmarks
BenchmarkDotNet
Ratio is relative to FastEndpoints. Lower is better.
| Endpoint | Mean | Ratio | Gen0 | Allocated | Alloc Ratio |
|---|---|---|---|---|---|
| MinimalApi | 19.84 μs | 0.87 | 0.7324 | 15.12 KB | 0.99 |
| FastEndpoints | 22.95 μs | 1.01 | 0.7324 | 15.25 KB | 1.00 |
| MVC Controller | 36.78 μs | 1.61 | 1.2207 | 22.52 KB | 1.48 |
NBomber Throughput Load Tests
hardware: AMD Ryzen 9 5950X (16c/32t), 32GB RAM
software: .NET 10.0 (CachyOS)
payload: same JSON payload as BenchmarkDotNet
load model: closed workload with 8 concurrent users
duration: 5 second warm-up, 1 minute testThroughput ratio is relative to FastEndpoints. Higher is better for requests/sec and throughput ratio. Lower is better for latency.
| Endpoint | Successful Requests | Requests/sec | Throughput Ratio | Mean Latency | P95 Latency | P99 Latency |
|---|---|---|---|---|---|---|
| Minimal APIs | 12,679,923 | 211,332.05 | 1.03 | 0.04 ms | 0.03 ms | 0.05 ms |
| FastEndpoints | 12,250,861 | 204,181.02 | 1.00 | 0.04 ms | 0.04 ms | 0.05 ms |
| MVC Controller | 8,678,790 | 144,646.50 | 0.71 | 0.05 ms | 0.05 ms | 0.90 ms |
TechEmpower Benchmark (Preliminary)
