Coi Benchmarks

Benchmark suites for Coi and related ecosystem performance tracking.

Goals

• Keep benchmark categories isolated by suite
• Support adding new benchmark types beyond framework comparisons
• Standardize execution and result output layout

Running Benchmarks

./run.sh                      # Runs default suite: framework
./run.sh framework            # Runs framework suite explicitly
./run.sh framework --dom-only # Pass suite-specific options through

Current Suites

• framework — Coi vs React vs Vue vs Svelte bundle and DOM benchmarks

Repository Structure

benchmarks/
├── run.sh                    # Suite orchestrator
└── suites/
	└── framework/
		├── apps/
		│   ├── counter/
		│   │   ├── coi/
		│   │   ├── react/
		│   │   ├── svelte/
		│   │   └── vue/
		│   └── rows/
		│       ├── coi/
		│       ├── react/
		│       ├── svelte/
		│       └── vue/
		├── benchmark.py
		├── results/
		│   ├── benchmark_results.json
		│   └── benchmark_results.svg
		├── README.md
		└── run.sh

Adding a New Benchmark Type

1. Create suites/<name>/
2. Add a run.sh entry script for that suite
3. Keep suite-specific apps/data under that suite directory
4. Write outputs to suites/<name>/results/
5. Document usage in suites/<name>/README.md