Skip to content

chore: update benchmark scripts for reporting and test runs #187

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
solderzzc merged 2 commits into from
Apr 8, 2026
+17 −0
Merged

chore: update benchmark scripts for reporting and test runs #187

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
92aad77
chore: update benchmark scripts for reporting and test runs
solderzzc Apr 8, 2026
17e4eab
Merge branch 'develop' into feature/coral-tpu-detection
solderzzc Apr 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions skills/analysis/home-security-benchmark/scripts/generate-report.cjs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,6 +109,7 @@ function buildHTML(allResults, fixtureImages, { liveMode = false, liveStatus = n
tokens: r.tokens || r.data?.tokenTotals?.total,
perfSummary: r.perfSummary || r.data?.perfSummary || null,
system: r.data?.system || {},
serverParams: r.data?.serverParams || {},
tokenTotals: r.data?.tokenTotals || {},
suites: (r.data?.suites || []).map(s => ({
name: s.name,
Expand Down Expand Up @@ -492,6 +493,15 @@ function renderPerformance() {
let html = '<div class="header"><div class="page-title">⚡ Performance</div>';
html += '<div class="page-subtitle">' + esc(run.model || '?') + ' — ' + shortDate(run.timestamp) + '</div></div>';

if (run.serverParams && typeof run.serverParams === 'object' && Object.keys(run.serverParams).length > 0) {
let paramStr = '';
for (const k in run.serverParams) {
if (paramStr) paramStr += ' | ';
paramStr += '<b>' + esc(k) + '</b>: ' + esc(String(run.serverParams[k]));
}
html += '<div style="font-size:0.75rem; color:var(--text-dim); margin-left: 2rem; margin-top: 0.5rem; padding-bottom: 0.5rem;">[Server Params] ' + paramStr + '</div>';
}

// Hero cards
html += '<div class="hero-grid">';
const ttftAvg = perf?.ttft?.avgMs;
Expand Down
7 changes: 7 additions & 0 deletions skills/analysis/home-security-benchmark/scripts/run-benchmark.cjs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -230,10 +230,14 @@ function suite(name, fn) {
suites.push({ name, fn, tests: [] });
}

let targetServerParams = {};
try { targetServerParams = JSON.parse(process.env.AEGIS_SERVER_PARAMS || '{}'); } catch { }

const results = {
timestamp: new Date().toISOString(),
gateway: GATEWAY_URL,
vlm: VLM_URL || null,
serverParams: targetServerParams,
system: {},
model: {},
suites: [],
Expand Down Expand Up @@ -333,6 +337,9 @@ async function llmCall(messages, opts = {}) {
...(model && { model }),
...(temperature !== undefined && { temperature }),
...(opts.expectJSON && { top_p: 0.8 }),
// For JSON-expected tests on local servers, enable server-side JSON mode
// which activates prefix buffering to strip hallucinated artifacts
...(opts.expectJSON && !isCloudApi && { response_format: { type: 'json_object' } }),
...(opts.tools && { tools: opts.tools }),
// Model-family-specific params (e.g. reasoning_effort:'none' for Mistral).
// These are merged last so they take precedence over defaults.
Expand Down