{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreicnvp3wrgvirojv7itdbtsvh5ol6ejlfautyjqsoax3czs3kyz2yi",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mica5qwjdjj2"
  },
  "path": "/t/atom-aipi-weekly-week-12-march-30-2026/174817#post_1",
  "publishedAt": "2026-03-30T15:55:06.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "Week 12 numbers are out.\n\n2,614 SKUs across 1,234 models and 47 vendors. All 14 AIPI indexes came in flat this week, 0.0% WoW across every modality and channel category.\n\nNumbers worth knowing:\n\nText input benchmark (AIPI TXT GLB): $0.001190 per 1K tokens\n\nText output benchmark: $0.005540 per 1K tokens\n\nOutput premium over input: 3.85x\n\nOpen source vs proprietary median: OSS is 70.7% cheaper\n\nCaching savings where available: 69.7%\n\nModels offering cached pricing: 20.4%\n\nContext cost curve (128K+ vs 32K): 3.0x\n\nThree straight weeks of flat pricing. The competition right now is on model breadth, not price.\n\nWe publish a free MCP server with four tools that work out of the box in Claude, Cursor, and Windsurf with no API key. MCP PRO at $49/mo unlocks search_models, get_model_detail, compare_prices, and get_vendor_catalog.",
  "title": "ATOM AIPI Weekly, Week 12 (March 30, 2026)"
}