Run Benchmark #10

	name: Run Benchmark

	on:
	workflow_dispatch:
	inputs:
	run:
	description: "Number of times to run the benchmark"
	required: true
	type: string
	model:
	description: "Model to use"
	required: true
	type: string
	tasks:
	description: "Comma-separated list of tasks"
	required: true
	type: string

	jobs:
	prepare:
	runs-on: ubuntu-latest
	outputs:
	tasks: ${{ steps.split.outputs.tasks }}
	steps:
	- name: Split tasks into matrix
	id: split
	run: \|
	TASKS_JSON=$(echo "${{ inputs.tasks }}" \| tr ',' '\n' \| sed 's/^[[:space:]]//;s/[[:space:]]$//' \| jq -R -s -c 'split("\n") \| map(select(length > 0))')
	echo "tasks=$TASKS_JSON" >> $GITHUB_OUTPUT

	benchmark:
	needs: prepare
	runs-on: ubuntu-latest
	strategy:
	matrix:
	task: ${{ fromJson(needs.prepare.outputs.tasks) }}
	environment: production
	steps:
	- name: Checkout repository
	uses: actions/checkout@v4

	- name: Setup Bun
	uses: oven-sh/setup-bun@v1
	with:
	bun-version: 1.2.21

	- name: Install dependencies
	run: bun install

	- name: Install OpenCode CLI
	run: bun add -g opencode-ai

	- name: Print benchmark config
	env:
	RUN_COUNT: ${{ inputs.run }}
	MODEL: ${{ inputs.model }}
	TASK: ${{ matrix.task }}
	run: \|
	echo "Run count: ${RUN_COUNT}"
	echo "Model: ${MODEL}"
	echo "Task: ${TASK}"

	- name: Run benchmark
	env:
	OPENCODE_API_KEY: ${{ secrets.OPENCODE_API_KEY }}
	DEBUG: true
	run: bun dev opencode --task ${{ matrix.task }} --model ${{ inputs.model }}

Provide feedback