Benchmark results (Intentum.Benchmarks)
This file is generated by running ./scripts/run-benchmarks.sh from the repository root.
Commit this file after running the script to keep published results in sync.
BenchmarkDotNet v0.15.8, macOS Tahoe 26.2 (25C56) [Darwin 25.2.0]
Apple M4 Pro, 1 CPU, 14 logical and 14 physical cores
.NET SDK 10.0.102
[Host] : .NET 10.0.2 (10.0.2, 10.0.225.61305), Arm64 RyuJIT armv8.0-a
DefaultJob : .NET 10.0.2 (10.0.2, 10.0.225.61305), Arm64 RyuJIT armv8.0-a
| Method | Mean | Error | StdDev | Gen0 | Gen1 | Allocated |
|---|---|---|---|---|---|---|
| ToVector_10Events | 0.2650 ns | 0.0052 ns | 0.0049 ns | - | - | - |
| ToVector_1KEvents | 0.2544 ns | 0.0029 ns | 0.0027 ns | - | - | - |
| ToVector_10KEvents | 0.3319 ns | 0.0033 ns | 0.0029 ns | - | - | - |
| LlmIntentModel_Infer_10Events | 273.7592 ns | 0.8363 ns | 0.7413 ns | 0.1326 | - | 1112 B |
| LlmIntentModel_Infer_1KEvents | 1,695.3022 ns | 8.4526 ns | 7.9066 ns | 0.5703 | 0.0038 | 4784 B |
| PolicyEngine_Decide | 12.8937 ns | 0.0204 ns | 0.0181 ns | 0.0029 | - | 24 B |
See Benchmarks for how to run and filter benchmarks. For improvement ideas and suggested solutions, see Benchmarks — Improvement opportunities.