jaakko/agentic-studio
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Benchmark: mallikohtainen yhteenvetotaulu + kokonaisaika

Browse Source 
Näyttää per malli: testit ja aika per skenaario, kokonaisläpäisy,
kokonaisaika, keskimääräinen tok/s ja keskipisteet.
This commit is contained in:
jaakko
2026-04-14 08:19:27 +03:00
parent 9f2899b83d
commit b3abf4e89f
1 changed files with 44 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

45
network-poc/tests/model-benchmark.mjs
View File

@@ -495,6 +495,48 @@ async function main() {
}
console.log('╚' + '═'.repeat(header.length + 2) + '╝');
// === Mallikohtainen yhteenveto ===
const modelNames = [...new Set(results.map(r => r.model))];
const scenarioIds = scenarios.map(s => s.id);
console.log('\n');
const mHeader = [
'Malli'.padEnd(35),
...scenarioIds.map(s => s.padEnd(18)),
'Yht.'.padEnd(8),
'Aika'.padEnd(8),
'tok/s'.padEnd(7),
'Pisteet',
].join(' │ ');
console.log(mHeader);
console.log('─'.repeat(mHeader.length));
for (const model of modelNames) {
const mrs = results.filter(r => r.model === model);
const cols = scenarioIds.map(sid => {
const r = mrs.find(r => r.scenario === sid);
if (!r) return '-'.padEnd(18);
const t = r.testsTotal > 0 ? `${r.testsPassed}/${r.testsTotal}` : '-';
const s = `${(r.totalDurationMs/1000).toFixed(0)}s`;
return `${t} ${s}`.padEnd(18);
});
const totalPassed = mrs.reduce((s, r) => s + r.testsPassed, 0);
const totalTests = mrs.reduce((s, r) => s + r.testsTotal, 0);
const totalTime = mrs.reduce((s, r) => s + r.totalDurationMs, 0);
const avgSpeed = mrs.length > 0 ? Math.round(mrs.reduce((s, r) => s + r.avgTokPerSec, 0) / mrs.length) : 0;
const avgScoreModel = mrs.length > 0 ? Math.round(mrs.reduce((s, r) => s + r.score, 0) / mrs.length) : 0;
const pct = totalTests > 0 ? Math.round(totalPassed / totalTests * 100) : 0;
const row = [
model.padEnd(35),
...cols,
`${totalPassed}/${totalTests}`.padEnd(8),
`${(totalTime/1000).toFixed(0)}s`.padEnd(8),
`${avgSpeed}`.padEnd(7),
`${starsForScore(avgScoreModel)} ${avgScoreModel}p (${pct}%)`,
].join(' │ ');
console.log(row);
}
// Tallenna JSON
writeFileSync(`${OUTPUT_DIR}/results.json`, JSON.stringify(results, null, 2));
console.log(`\nJSON: ${OUTPUT_DIR}/results.json`);
@@ -504,7 +546,8 @@ async function main() {
const partial = results.filter(r => !r.error && r.testsPassed < r.testsTotal && r.testsTotal > 0);
const failed = results.filter(r => r.error || r.testsTotal === 0);
const avgScore = results.length > 0 ? Math.round(results.reduce((s, r) => s + r.score, 0) / results.length) : 0;
console.log(`\n${starsForScore(avgScore)} Keskiarvo: ${avgScore}p | ✓ PASS: ${passed.length} | ◐ PARTIAL: ${partial.length} | ✗ FAIL: ${failed.length} | Yhteensä: ${results.length}`);
const totalTime = results.reduce((s, r) => s + r.totalDurationMs, 0);
console.log(`\n${starsForScore(avgScore)} Keskiarvo: ${avgScore}p | ✓ PASS: ${passed.length} | ◐ PARTIAL: ${partial.length} | ✗ FAIL: ${failed.length} | Yhteensä: ${results.length} | Kokonaisaika: ${(totalTime/1000/60).toFixed(1)} min`);
}
main().catch(e => { console.error(e); process.exit(1); });