{
  "_notice": {
    "type": "free_key_offer",
    "message": "Free and anonymous access is staying. If this endpoint is in your pipeline, a free key adds advance notice before any schema change or deprecation. POST /v1/keys {\"email\": \"you@firm.com\"} — nothing else required. Something broken or missing? hello@clearmarket.fyi reaches a human.",
    "docs": "https://clearmarket.fyi/for-data/",
    "policy": "This field is additive and its shape will not change."
  },
  "event_id": "CM-EVT-3ZGYWKCX43",
  "slug": "highest-claude-score-on-humanitys-last-exam-in-2026-20260723190836285",
  "question": "Anthropic Claude model highest score on Humanity's Last Exam, 2026",
  "category": "technology",
  "event_type": "LADDER",
  "ladder_distribution": {
    "direction": "above",
    "implied_band": [
      60,
      65
    ],
    "strikes": [
      {
        "threshold": 55,
        "prob": 84
      },
      {
        "threshold": 60,
        "prob": 50
      },
      {
        "threshold": 65,
        "prob": 26
      },
      {
        "threshold": 70,
        "prob": 8
      },
      {
        "threshold": 75,
        "prob": 3
      }
    ]
  },
  "tags": [
    "technology",
    "ai",
    "anthropic-claude",
    "benchmark-performance",
    "tech",
    "2026"
  ],
  "catalyst_types": [],
  "catalyst_dates": [],
  "editorial_notes": "These markets track the peak score any Anthropic Claude model achieves on Humanity's Last Exam by December 31, 2026. All five are Polymarket binary strikes at 55%, 60%, 65%, 70%, and 75% thresholds. Higher-strike markets are structurally subordinate: a YES at 70% implies YES at all lower strikes. Resolution references the official HLE leaderboard at agi.safe.ai via Polymarket's UMA Optimistic Oracle.",
  "venues_covered": [
    "polymarket"
  ],
  "primary_market_id": "CM-MKT-HX5NH0PMW0",
  "current_primary_mark": {
    "last_price": 0.26,
    "implied_probability": 0.26
  },
  "created_at": "2026-08-04T20:03:51.865287+00:00",
  "updated_at": "2026-08-04T20:03:51.865287+00:00",
  "markets": [
    {
      "market_id": "CM-MKT-H10SFPDWB3",
      "event_id": "CM-EVT-3ZGYWKCX43",
      "platform": "polymarket",
      "platform_market_id": "0x473068c53ac55ec2ac283a460f729961fd52946ece83cb2115bbb8297471bb9d",
      "question": "Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 55% or higher?",
      "description": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable dur",
      "contract_type": "binary",
      "settlement_currency": "USDC",
      "underlying_reference": "55%+: The official Humanity's Last Exam leaderboard at agi.safe.ai resolves whether the named model achieved at least the specified accuracy threshold by December 31, 2026, 11:59 PM ET.",
      "close_at": "2026-12-31T23:59:00Z",
      "resolve_at": "2026-12-31T23:59:00Z",
      "status": "open",
      "resolution": {
        "rules_raw": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to \"No\".",
        "arbitration_model": "uma_oracle",
        "proposer": "managed_whitelist",
        "source": "https://agi.safe.ai/",
        "source_status": "platform_named",
        "source_citation": "https://agi.safe.ai/",
        "sources": [
          {
            "name": "https://agi.safe.ai/",
            "url": null,
            "provenance": "platform_api"
          },
          {
            "name": "agi.safe.ai",
            "url": "https://agi.safe.ai/",
            "provenance": "platform_api"
          }
        ],
        "source_of_record": "https://agi.safe.ai/",
        "source_mechanism": "precedence",
        "source_type": null,
        "source_quality": null,
        "source_commitment": "named",
        "source_commitment_subtype": "named",
        "source_hedge_text": null
      },
      "rcg": {
        "grade": "B",
        "score": 66,
        "caps": [],
        "applied_factors": 6
      },
      "last_price": 0.844,
      "implied_probability": 0.844,
      "volume_24h_usd": 50,
      "volume_total_usd": 5517.513437,
      "settlement_style": "terminal",
      "direction": "above",
      "threshold": 55,
      "question_id": "CMX-D7B86B1417",
      "also_on": null,
      "tags": [
        "ai benchmark",
        "anthropic",
        "claude",
        "claude highest score",
        "humanity's last exam"
      ],
      "eligibility_screens": null
    },
    {
      "market_id": "CM-MKT-ZKSVXR7870",
      "event_id": "CM-EVT-3ZGYWKCX43",
      "platform": "polymarket",
      "platform_market_id": "0x3fb3dfa7e6332d421dfb49d735e05b4de1a0f9b7af42d3e4b22f56dfcd1f2858",
      "question": "Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 60% or higher?",
      "description": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable dur",
      "contract_type": "binary",
      "settlement_currency": "USDC",
      "underlying_reference": "60%+: The official Humanity's Last Exam leaderboard at agi.safe.ai resolves whether the named model achieved at least the specified accuracy threshold by December 31, 2026, 11:59 PM ET.",
      "close_at": "2026-12-31T23:59:00Z",
      "resolve_at": "2026-12-31T23:59:00Z",
      "status": "open",
      "resolution": {
        "rules_raw": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to \"No\".",
        "arbitration_model": "uma_oracle",
        "proposer": "managed_whitelist",
        "source": "https://agi.safe.ai/",
        "source_status": "platform_named",
        "source_citation": "https://agi.safe.ai/",
        "sources": [
          {
            "name": "https://agi.safe.ai/",
            "url": null,
            "provenance": "platform_api"
          },
          {
            "name": "agi.safe.ai",
            "url": "https://agi.safe.ai/",
            "provenance": "platform_api"
          }
        ],
        "source_of_record": "https://agi.safe.ai/",
        "source_mechanism": "precedence",
        "source_type": null,
        "source_quality": null,
        "source_commitment": "named",
        "source_commitment_subtype": "named",
        "source_hedge_text": null
      },
      "rcg": {
        "grade": "B",
        "score": 66,
        "caps": [],
        "applied_factors": 6
      },
      "last_price": 0.5,
      "implied_probability": 0.5,
      "volume_24h_usd": 206.18,
      "volume_total_usd": 13835.447854999999,
      "settlement_style": "terminal",
      "direction": "above",
      "threshold": 60,
      "question_id": "CMX-32D5EACD9D",
      "also_on": null,
      "tags": [
        "ai benchmark",
        "anthropic",
        "claude",
        "claude highest score",
        "humanity's last exam"
      ],
      "eligibility_screens": null
    },
    {
      "market_id": "CM-MKT-HX5NH0PMW0",
      "event_id": "CM-EVT-3ZGYWKCX43",
      "platform": "polymarket",
      "platform_market_id": "0xfa55e7e4900a3c2a0b09834ec7fb505ada92e0c0d39bb738a5c3a4357a64316f",
      "question": "Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 65% or higher?",
      "description": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable dur",
      "contract_type": "binary",
      "settlement_currency": "USDC",
      "underlying_reference": "65%+: The official Humanity's Last Exam leaderboard at agi.safe.ai resolves whether the named model achieved at least the specified accuracy threshold by December 31, 2026, 11:59 PM ET.",
      "close_at": "2026-12-31T23:59:00Z",
      "resolve_at": "2026-12-31T23:59:00Z",
      "status": "open",
      "resolution": {
        "rules_raw": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to \"No\".",
        "arbitration_model": "uma_oracle",
        "proposer": "managed_whitelist",
        "source": "https://agi.safe.ai/",
        "source_status": "platform_named",
        "source_citation": "https://agi.safe.ai/",
        "sources": [
          {
            "name": "https://agi.safe.ai/",
            "url": null,
            "provenance": "platform_api"
          },
          {
            "name": "agi.safe.ai",
            "url": "https://agi.safe.ai/",
            "provenance": "platform_api"
          }
        ],
        "source_of_record": "https://agi.safe.ai/",
        "source_mechanism": "precedence",
        "source_type": null,
        "source_quality": null,
        "source_commitment": "named",
        "source_commitment_subtype": "named",
        "source_hedge_text": null
      },
      "rcg": {
        "grade": "B",
        "score": 66,
        "caps": [],
        "applied_factors": 6
      },
      "last_price": 0.26,
      "implied_probability": 0.26,
      "volume_24h_usd": null,
      "volume_total_usd": 9905.351177,
      "settlement_style": "terminal",
      "direction": "above",
      "threshold": 65,
      "question_id": "CMX-AA94B64A1D",
      "also_on": null,
      "tags": [
        "ai benchmark",
        "anthropic",
        "claude",
        "claude highest score",
        "humanity's last exam"
      ],
      "eligibility_screens": null
    },
    {
      "market_id": "CM-MKT-P0T3KWK9B9",
      "event_id": "CM-EVT-3ZGYWKCX43",
      "platform": "polymarket",
      "platform_market_id": "0xddfa4331afe2cd3e4058f67f08404652f80a48e4792fa57f4284b7e3be2e3572",
      "question": "Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 70% or higher?",
      "description": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable dur",
      "contract_type": "binary",
      "settlement_currency": "USDC",
      "underlying_reference": "70%+: The official Humanity's Last Exam leaderboard at agi.safe.ai resolves whether the named model achieved at least the specified accuracy threshold by December 31, 2026, 11:59 PM ET.",
      "close_at": "2026-12-31T23:59:00Z",
      "resolve_at": "2026-12-31T23:59:00Z",
      "status": "open",
      "resolution": {
        "rules_raw": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to \"No\".",
        "arbitration_model": "uma_oracle",
        "proposer": "managed_whitelist",
        "source": "https://agi.safe.ai/",
        "source_status": "platform_named",
        "source_citation": "https://agi.safe.ai/",
        "sources": [
          {
            "name": "https://agi.safe.ai/",
            "url": null,
            "provenance": "platform_api"
          },
          {
            "name": "agi.safe.ai",
            "url": "https://agi.safe.ai/",
            "provenance": "platform_api"
          }
        ],
        "source_of_record": "https://agi.safe.ai/",
        "source_mechanism": "precedence",
        "source_type": null,
        "source_quality": null,
        "source_commitment": "named",
        "source_commitment_subtype": "named",
        "source_hedge_text": null
      },
      "rcg": {
        "grade": "B",
        "score": 66,
        "caps": [],
        "applied_factors": 6
      },
      "last_price": 0.08,
      "implied_probability": 0.08,
      "volume_24h_usd": 70,
      "volume_total_usd": 1860.8200000000002,
      "settlement_style": "terminal",
      "direction": "above",
      "threshold": 70,
      "question_id": "CMX-28D51EA70D",
      "also_on": null,
      "tags": [
        "ai benchmark",
        "anthropic",
        "claude",
        "claude highest score",
        "humanity's last exam"
      ],
      "eligibility_screens": null
    },
    {
      "market_id": "CM-MKT-WWL7556H77",
      "event_id": "CM-EVT-3ZGYWKCX43",
      "platform": "polymarket",
      "platform_market_id": "0x0ec854f7b18a669ebf06b8426cc69d6a7cfbaae674096e19e46fdd76f1837187",
      "question": "Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 75% or higher?",
      "description": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable dur",
      "contract_type": "binary",
      "settlement_currency": "USDC",
      "underlying_reference": "75%+: The official Humanity's Last Exam leaderboard at agi.safe.ai resolves whether the named model achieved at least the specified accuracy threshold by December 31, 2026, 11:59 PM ET.",
      "close_at": "2026-12-31T23:59:00Z",
      "resolve_at": "2026-12-31T23:59:00Z",
      "status": "open",
      "resolution": {
        "rules_raw": "This market will resolve to \"Yes\" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to \"No\".\n\nFor resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.\n\nThe resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to \"No\".",
        "arbitration_model": "uma_oracle",
        "proposer": "managed_whitelist",
        "source": "https://agi.safe.ai/",
        "source_status": "platform_named",
        "source_citation": "https://agi.safe.ai/",
        "sources": [
          {
            "name": "https://agi.safe.ai/",
            "url": null,
            "provenance": "platform_api"
          },
          {
            "name": "agi.safe.ai",
            "url": "https://agi.safe.ai/",
            "provenance": "platform_api"
          }
        ],
        "source_of_record": "https://agi.safe.ai/",
        "source_mechanism": "precedence",
        "source_type": null,
        "source_quality": null,
        "source_commitment": "named",
        "source_commitment_subtype": "named",
        "source_hedge_text": null
      },
      "rcg": {
        "grade": "B",
        "score": 66,
        "caps": [],
        "applied_factors": 6
      },
      "last_price": 0.03,
      "implied_probability": 0.03,
      "volume_24h_usd": 20,
      "volume_total_usd": 901.026061,
      "settlement_style": "terminal",
      "direction": "above",
      "threshold": 75,
      "question_id": "CMX-8DDB5541A9",
      "also_on": null,
      "tags": [
        "ai benchmark",
        "anthropic",
        "claude",
        "claude highest score",
        "humanity's last exam"
      ],
      "eligibility_screens": null
    }
  ],
  "resolution_log": [],
  "_provenance": {
    "source": "clearmarket.fyi",
    "ref": "CM-EVT-3ZGYWKCX43",
    "terms": "Attribution required. clearmarket.fyi"
  }
}