nsanku-mmlu-leaderboard

Running

File size: 11,896 Bytes

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Ghanaian MMLU Leaderboard</title>
<style>
*, *::before, *::after { box-sizing: border-box; margin: 0; padding: 0; }
:root {
  --bg: #f8fafc; --surface: #ffffff; --surface-2: #f1f5f9; --border: #e2e8f0;
  --text: #1e293b; --text-muted: #64748b; --accent: #4f46e5; --accent-hover: #4338ca;
  --success: #059669; --danger: #dc2626; --warning: #d97706;
  --table-stripe: rgba(15,23,42,0.025); --radius: 12px;
}
html { font-size: 14px; }
body { font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, 'Helvetica Neue', Arial, sans-serif;
  background: var(--bg); color: var(--text); line-height: 1.5; min-height: 100vh; }
a { color: var(--accent-hover); text-decoration: none; }
a:hover { text-decoration: underline; }
.container { max-width: 1200px; margin: 0 auto; padding: 24px 20px 60px; }
.header { text-align: center; margin-bottom: 28px; }
.header h1 { font-size: 1.8rem; font-weight: 700; margin-bottom: 8px; }
.header p { color: var(--text-muted); font-size: 0.95rem; }
.header .links { margin-top: 10px; }
.repo-link { display: inline-flex; align-items: center; gap: 4px; font-size: 0.85rem; }
.repo-link svg { width: 16px; height: 16px; }
.tabs { display: flex; gap: 4px; margin-bottom: 24px; flex-wrap: wrap; }
.tab { padding: 10px 20px; border-radius: 8px 8px 0 0; cursor: pointer;
  border: 1px solid var(--border); background: var(--surface); color: var(--text-muted);
  font-weight: 500; font-size: 0.95rem; transition: all 0.15s; border-bottom: none; }
.tab:hover { color: var(--text); }
.tab.active { background: var(--surface-2); color: var(--text); border-bottom: 2px solid var(--accent); }
.tab-pane { display: none; }
.tab-pane.active { display: block; }
.stats-grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(160px, 1fr));
  gap: 12px; margin-bottom: 24px; }
.stat-card { background: var(--surface); border: 1px solid var(--border);
  border-radius: var(--radius); padding: 18px; }
.stat-card .label { color: var(--text-muted); font-size: 0.8rem; margin-bottom: 6px; }
.stat-card .value { font-size: 1.4rem; font-weight: 700; }
.table-wrap { background: var(--surface); border: 1px solid var(--border);
  border-radius: var(--radius); overflow-x: auto; }
table { width: 100%; border-collapse: collapse; font-size: 0.88rem; }
th, td { text-align: left; padding: 10px 16px; white-space: nowrap; }
th { background: var(--surface-2); color: var(--text-muted); font-weight: 600; font-size: 0.8rem;
  text-transform: uppercase; letter-spacing: 0.05em; border-bottom: 1px solid var(--border);
  position: sticky; top: 0; }
tbody tr { border-bottom: 1px solid var(--border); }
tbody tr:nth-child(even) { background: var(--table-stripe); }
tbody tr:hover { background: rgba(79, 70, 229, 0.1); }
.empty-row { text-align: center; color: var(--text-muted); }
.mtype { display: inline-block; padding: 1px 8px; border-radius: 4px; font-size: 0.75rem;
  background: var(--surface-2); color: var(--text-muted); }
.rank-medal { font-size: 1rem; }
::-webkit-scrollbar { width: 8px; height: 8px; }
::-webkit-scrollbar-track { background: var(--surface); }
::-webkit-scrollbar-thumb { background: var(--border); border-radius: 4px; }
.loading { text-align: center; padding: 80px 0; color: var(--text-muted); }
.loading .spinner { width: 40px; height: 40px; border: 3px solid var(--border);
  border-top-color: var(--accent); border-radius: 50%; animation: spin 0.8s linear infinite; margin: 0 auto 16px; }
@keyframes spin { to { transform: rotate(360deg); } }
.footer { margin-top: 40px; text-align: center; color: var(--text-muted); font-size: 0.8rem;
  padding-top: 20px; border-top: 1px solid var(--border); }
.note { color: var(--text-muted); margin-bottom: 14px; font-size: 0.9rem; }
@media (max-width: 600px) {
  .header h1 { font-size: 1.4rem; }
  th, td { padding: 8px 10px; font-size: 0.82rem; }
}
</style>
</head>
<body>
<div class="container">
  <div class="header">
    <h1>🇬🇭 Ghanaian MMLU Leaderboard</h1>
    <p>Benchmarking LLMs on <strong>2,194</strong> multiple-choice questions across
       <strong>12 Ghanaian languages</strong> from the
       <a href="https://ntc.gov.gh/practice_test/ghanaian_languages/" target="_blank">NTC licensing practice tests</a>.
       All models evaluated with <strong>thinking disabled</strong> for fair comparison.</p>
    <div class="links">
      <a href="https://github.com/GhanaNLP/nsanku-MMLU" target="_blank" class="repo-link">
        <svg viewBox="0 0 16 16" fill="currentColor"><path d="M8 0C3.58 0 0 3.58 0 8c0 3.54 2.29 6.53 5.47 7.59.4.07.55-.17.55-.38 0-.19-.01-.82-.01-1.49-2.01.37-2.53-.49-2.69-.94-.09-.23-.48-.94-.82-1.13-.28-.15-.68-.52-.01-.53.63.01 1.08.58 1.23.82.72 1.21 1.87.87 2.33.66.07-.52.28-.87.51-1.07-1.78-.2-3.64-.89-3.64-3.95 0-.87.31-1.59.82-2.15-.08-.2-.36-1.02.08-2.12 0 0 .67-.21 2.2.82.64-.18 1.32-.27 2-.27.68 0 1.36.09 2 .27 1.53-1.04 2.2-.82 2.2-.82.44 1.1.16 1.92.08 2.12.51.56.82 1.27.82 2.15 0 3.07-1.87 3.75-3.65 3.95.29.25.54.73.54 1.48 0 1.07-.01 1.93-.01 2.2 0 .21.15.46.55.38A8.013 8.013 0 0016 8c0-4.42-3.58-8-8-8z"/></svg>
        GhanaNLP/nsanku-MMLU
      </a>
    </div>
  </div>

  <div id="loading" class="loading">
    <div class="spinner"></div>
    <p>Loading benchmark data from GitHub…</p>
  </div>

  <div id="main" style="display:none;">
    <div class="stats-grid" id="stats-grid"></div>

    <div class="tabs">
      <div class="tab active" data-tab="overall">🏆 Overall</div>
      <div class="tab" data-tab="per-lang">🌐 Per-Language</div>
    </div>

    <div class="tab-pane active" id="pane-overall">
      <p class="note">Overall accuracy across all 2,194 questions — higher is better.</p>
      <div class="table-wrap">
        <table>
          <thead><tr><th>#</th><th>Model</th><th>Accuracy (%)</th><th>Correct</th><th>Total</th><th>Unparsed</th></tr></thead>
          <tbody id="overall-body"></tbody>
        </table>
      </div>
    </div>

    <div class="tab-pane" id="pane-per-lang">
      <p class="note">Per-language accuracy (%) — models sorted by overall accuracy.</p>
      <div class="table-wrap">
        <table>
          <thead><tr id="per-lang-head"></tr></thead>
          <tbody id="per-lang-body"></tbody>
        </table>
      </div>
    </div>

    <div class="footer">
      <p>Higher accuracy is better · Chance baseline ≈ 25% (4 options)</p>
      <p>Add a model by opening a PR to <a href="https://github.com/GhanaNLP/nsanku-MMLU" target="_blank">the GitHub repo</a> — see CONTRIBUTING.md. Then reload this page.</p>
    </div>
  </div>
</div>

<script>
const REPO = 'GhanaNLP/nsanku-MMLU';
const BRANCH = 'master';
const RAW = `https://raw.githubusercontent.com/${REPO}/${BRANCH}`;
const API = `https://api.github.com/repos/${REPO}/contents`;

const DISPLAY_NAMES = {
  'openai/gpt-4o': 'GPT-4o',
  'gemini/gemini-3-flash-preview': 'Gemini 3 Flash',
  'gemini/gemini-3.1-pro-preview': 'Gemini 3.1 Pro',
  'nvidia/deepseek-ai/deepseek-v4-flash': 'DeepSeek v4 Flash',
  'nvidia/minimaxai/minimax-m3': 'MiniMax M3',
  'nvidia/z-ai/glm-5.2': 'GLM-5.2',
  'nvidia/nvidia/nemotron-3-ultra-550b-a55b': 'Nemotron 3 Ultra 550B',
  'nvidia/abacusai/dracarys-llama-3.1-70b-instruct': 'Dracarys Llama 3.1 70B',
  'mistral/mistral-large-latest': 'Mistral Large',
};
const LANG_ORDER = ['akuapem_twi','asante_twi','dagaare','dagbani','dangme','ewe','fanti','ga','gonja','gurene','kasem','nzema'];
const LANG_DISPLAY = {akuapem_twi:'Akuapem Twi',asante_twi:'Asante Twi',dagaare:'Dagaare',dagbani:'Dagbani',
  dangme:'Dangme',ewe:'Ewe',fanti:'Fante',ga:'Ga',gonja:'Gonja',gurene:'Gurene',kasem:'Kasem',nzema:'Nzema'};

function displayName(spec){ return DISPLAY_NAMES[spec] || (spec||'unknown').split('/').pop(); }
function escapeHtml(s){ return (s||'').replace(/&/g,'&amp;').replace(/</g,'&lt;').replace(/>/g,'&gt;'); }
function accColor(a){ return a >= 50 ? 'var(--success)' : a >= 35 ? 'var(--accent-hover)' : 'var(--warning)'; }
const MEDALS = ['🥇','🥈','🥉'];

async function getJSON(url){ const r = await fetch(url); if(!r.ok) throw new Error('HTTP '+r.status+' for '+url); return r.json(); }

async function loadResults(){
  const entries = await getJSON(`${API}/results?ref=${BRANCH}`);
  const dirs = entries.filter(e => e.type === 'dir');
  const out = [];
  for (const d of dirs){
    try {
      const data = await getJSON(`${RAW}/results/${d.name}/results.json`);
      if (data && data.overall) out.push(data);
    } catch(e){ /* dir without results.json (incomplete/dry-run) — skip */ }
  }
  return out;
}

function renderStats(results){
  const n = results.length;
  const best = results.reduce((a,b)=> b.overall.accuracy > (a?a.overall.accuracy:-1) ? b : a, null);
  const stats = [
    {label:'Models', value:n},
    {label:'Questions', value:'2,194'},
    {label:'Languages', value:'12'},
    {label:'Best model', value: best ? `${displayName(best.model)} · ${best.overall.accuracy.toFixed(2)}%` : '—'},
  ];
  document.getElementById('stats-grid').innerHTML = stats.map(s =>
    `<div class="stat-card"><div class="label">${s.label}</div><div class="value">${s.value}</div></div>`).join('');
}

function renderOverall(results){
  const rows = [...results].sort((a,b)=> b.overall.accuracy - a.overall.accuracy);
  const body = document.getElementById('overall-body');
  if(!rows.length){ body.innerHTML = '<tr class="empty-row"><td colspan="6">No results found</td></tr>'; return; }
  body.innerHTML = rows.map((r,i)=>{
    const o = r.overall;
    return `<tr>
      <td>${MEDALS[i] ? `<span class="rank-medal">${MEDALS[i]}</span>` : (i+1)}</td>
      <td>${escapeHtml(displayName(r.model))}</td>
      <td style="font-weight:700;color:${accColor(o.accuracy)}">${o.accuracy.toFixed(2)}</td>
      <td>${o.correct}</td>
      <td>${o.total}</td>
      <td>${o.unparsed||0}</td>
    </tr>`;
  }).join('');
}

function renderPerLang(results){
  const rows = [...results].sort((a,b)=> b.overall.accuracy - a.overall.accuracy);
  document.getElementById('per-lang-head').innerHTML =
    '<th>Model</th>' + LANG_ORDER.map(l=>`<th>${LANG_DISPLAY[l]}</th>`).join('');
  const body = document.getElementById('per-lang-body');
  if(!rows.length){ body.innerHTML = '<tr class="empty-row"><td colspan="13">No results found</td></tr>'; return; }
  body.innerHTML = rows.map(r=>{
    const cells = LANG_ORDER.map(l=>{
      const s = (r.per_language||{})[l];
      if(s && s.total>0) return `<td style="color:${accColor(s.accuracy)}">${s.accuracy.toFixed(1)}</td>`;
      return '<td>—</td>';
    }).join('');
    return `<tr><td>${escapeHtml(displayName(r.model))}</td>${cells}</tr>`;
  }).join('');
}

function setupTabs(){
  document.querySelectorAll('.tab').forEach(tab=>{
    tab.addEventListener('click', ()=>{
      document.querySelectorAll('.tab').forEach(t=>t.classList.remove('active'));
      document.querySelectorAll('.tab-pane').forEach(p=>p.classList.remove('active'));
      tab.classList.add('active');
      document.getElementById('pane-'+tab.dataset.tab).classList.add('active');
    });
  });
}

async function main(){
  try {
    const results = await loadResults();
    document.getElementById('loading').style.display = 'none';
    document.getElementById('main').style.display = 'block';
    renderStats(results);
    renderOverall(results);
    renderPerLang(results);
    setupTabs();
  } catch(e){
    document.getElementById('loading').innerHTML =
      `<div style="color:var(--danger)"><p>Failed to load benchmark data.</p>
       <p style="font-size:0.85rem;margin-top:8px">${escapeHtml(e.message)}</p>
       <p style="margin-top:12px;font-size:0.85rem">Visit the <a href="https://github.com/GhanaNLP/nsanku-MMLU" target="_blank">GitHub repo</a> directly.</p></div>`;
  }
}
main();
</script>
</body>
</html>