{
  "schema_version": 1,
  "generated_at_utc": "2026-11-01T17:38:52.550720+00:00",
  "data_class": "selected_development_examples",
  "comparable_aggregate": null,
  "leaderboard_ready": false,
  "selection": "One verified clear per distinct state model and selected verified pixels-only partial runs. Different development prompts, cadence and history. No best-of result is a model average.",
  "models_href": "models.json",
  "model_counts": {
    "state": 22,
    "image": 14
  },
  "tracks": [
    {
      "id": "state",
      "input": "current visible memory-derived state",
      "verified_clear_models": 4
    },
    {
      "id": "image",
      "input": "pixels plus declared controls and timing only; no RAM",
      "verified_clear_models": 1
    }
  ],
  "metrics": {
    "primary": {
      "id": "mean_progress_pct",
      "direction": "higher_is_better",
      "range": [
        0,
        100
      ],
      "aggregation": "mean all prespecified repeats within level; equal level weights; not available for selected examples"
    },
    "speedrun": {
      "id": "speedrun_score_native_frames",
      "direction": "lower_is_better",
      "completed_only": true,
      "nominal_seconds": "native_frames/60",
      "aggregation": "completed-only median per level with attempt/completion counts; unavailable for these selected examples",
      "failed_episode_value": null
    }
  },
  "levels": [
    {
      "level": "1-1",
      "start_x": 40,
      "goal_x": 3161,
      "progress_mapping_validated": true
    },
    {
      "level": "1-2",
      "start_x": 40,
      "goal_x": null,
      "progress_mapping_validated": false
    },
    {
      "level": "1-3",
      "start_x": 40,
      "goal_x": null,
      "progress_mapping_validated": false
    },
    {
      "level": "1-4",
      "start_x": 40,
      "goal_x": null,
      "progress_mapping_validated": false
    }
  ],
  "search_context": [
    {
      "scope": "first state development search",
      "episodes_started": 145,
      "episodes_finished": 135,
      "unfinished_or_interrupted": 10,
      "completed_episodes": 3,
      "distinct_completed_models": 2,
      "note": "Includes two Pplx clears and one Liquid clear; separate additional-state search not included"
    },
    {
      "scope": "additional pixels development search",
      "index_sha256": "d103bf6034a1735fbde07bf053fe4613b947281a3fcef11fe33dc30b15ad8e00",
      "note": "All attempted runs retained in local research registry; selected episodes are not a denominator for clear rate",
      "episodes_started": 92,
      "status_counts": {
        "death": 69,
        "development_stagnation_abort": 14,
        "api_failure": 2,
        "development_external_stagnation_abort": 2,
        "running": 4,
        "development_external_abort": 1
      },
      "completed_episodes": 0
    },
    {
      "scope": "first pixels development search",
      "episodes_started": 168,
      "episodes_finished": 151,
      "unfinished_or_interrupted": 17,
      "completed_episodes": 0,
      "note": "Directory snapshot of all exploratory profiles; not a matched cohort or clear-rate estimate"
    },
    {
      "scope": "additional state development search",
      "episodes_started": 131,
      "episodes_finished": 124,
      "unfinished_or_interrupted": 7,
      "completed_episodes": 4,
      "note": "Independent exploratory contracts and cadence sweeps; no best-of clear rate"
    }
  ],
  "examples": [
    {
      "id": "pplx-state-clear",
      "data_class": "selected_development_examples",
      "model_id": "perplexity/pplx-decider-v1.2",
      "display_name": "Perplexity: Decider V1.2",
      "returned_model": "perplexity/pplx-decider-v1.2-20261016",
      "provider": "Perplexity",
      "mode": "state",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3155.0,
      "progress_pct": 100.0,
      "native_frames": 1208,
      "decisions": 302,
      "speedrun_score_native_frames": 1208,
      "nominal_game_seconds_to_completion": 20.133333333333333,
      "experiment_config": {
        "profile": "phase guidance; compact visible state",
        "adapter": "joint12",
        "cadence": 4,
        "previous_screens": 0,
        "observation": "native current state and recent positions",
        "contract_sha256": "905208dc2a10e81fc31923f84ca614968e21979de9141a8f2409c238a4e63a3c",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "pplx-state-clear"
      },
      "input_audit": {
        "privileged_state_sent": true,
        "description": "state track intentionally receives current visible memory-derived state; independently replayed requests and outputs"
      },
      "trace_href": "traces/pplx-state-clear.json",
      "trace_sha256": "f2b470ca6ce60e07f60a69a36d1a965db44bbde75abc81da51038a6f5be02891",
      "replay_href": "replays/pplx-state-clear.mp4",
      "replay_sha256": "06949c052e4823c82b4736de0e67f074d7b532237dffbc2fc6f2fb723b5844b6",
      "source_hashes": {
        "episode_sha256": "e2e503b1067ec1e485c5e82cc88581868341a5544d9ffc8f794a92e12c0bbcdb"
      },
      "controls": {
        "adapter": "joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/pplx-state-clear.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 1,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/pplx-state-clear.json",
      "profile_sha256": "05b21e926579c535547784e784dc1608ce19cd61ecf677567eaddc3ce9c2a8ec"
    },
    {
      "id": "mercury-state-clear",
      "data_class": "selected_development_examples",
      "model_id": "inception/mercury-decide",
      "display_name": "Inception: Mercury Decide",
      "returned_model": "inception/mercury-decide-20260930",
      "provider": "Inception",
      "mode": "state",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3157.0,
      "progress_pct": 100.0,
      "native_frames": 1320,
      "decisions": 264,
      "speedrun_score_native_frames": 1320,
      "nominal_game_seconds_to_completion": 22.0,
      "experiment_config": {
        "profile": "phase guidance; compact visible state",
        "adapter": "joint12",
        "cadence": 5,
        "previous_screens": 0,
        "observation": "native current state and recent positions",
        "contract_sha256": "905208dc2a10e81fc31923f84ca614968e21979de9141a8f2409c238a4e63a3c",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "mercury-state-clear"
      },
      "input_audit": {
        "privileged_state_sent": true,
        "description": "state track intentionally receives current visible memory-derived state; independently replayed requests and outputs"
      },
      "trace_href": "traces/mercury-state-clear.json",
      "trace_sha256": "02125cb726ce2aa17fbd624f0ea5c0f94f1f832d560c510e9a99800de7271d34",
      "replay_href": "replays/mercury-state-clear.mp4",
      "replay_sha256": "343500fbd58d3c6233f90a8b7912166afdc451989d63111eebf545e7adb96b4f",
      "source_hashes": {
        "episode_sha256": "e6ccedd108f4145dcf8dc9eadee5440b9db53477fd4d814d353ef53303adc536"
      },
      "controls": {
        "adapter": "joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/mercury-state-clear.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 1,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/mercury-state-clear.json",
      "profile_sha256": "b3596ca64ff41529d76c47c817b1f9cedb0c5ad411e7119cb2714d0366c7e025"
    },
    {
      "id": "drex-state-clear",
      "data_class": "selected_development_examples",
      "model_id": "nace-ai/drex-max",
      "display_name": "Nace.AI: Drex Max",
      "returned_model": "nace-ai/drex-max-20261016",
      "provider": "DeepInfra",
      "mode": "state",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3156.0,
      "progress_pct": 100.0,
      "native_frames": 1384,
      "decisions": 116,
      "speedrun_score_native_frames": 1384,
      "nominal_game_seconds_to_completion": 23.066666666666666,
      "experiment_config": {
        "profile": "short rules; compact visible state",
        "adapter": "joint12",
        "cadence": 12,
        "previous_screens": 0,
        "observation": "native current state and recent positions",
        "contract_sha256": "44c3be57bf21acee2a4f004d9ec4b528b3a8d605d879bccb7a8ca5c3607dc7ee",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "drex-state-clear"
      },
      "input_audit": {
        "privileged_state_sent": true,
        "description": "state track intentionally receives current visible memory-derived state; independently replayed requests and outputs"
      },
      "trace_href": "traces/drex-state-clear.json",
      "trace_sha256": "0af10e40124efaea2def0e4c4e6339c5ae458a2f8f15212c2b10f837d278c6bc",
      "replay_href": "replays/drex-state-clear.mp4",
      "replay_sha256": "cdc6b96fbbc032ead6a85db0cf42076d5b32f1694d8ce58929df6cf93c971a6c",
      "source_hashes": {
        "episode_sha256": "dc0a8981e5e36618ee6cd8e08d55af740d621138f929a9144a45d80926f47fca"
      },
      "controls": {
        "adapter": "joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/drex-state-clear.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 1,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/drex-state-clear.json",
      "profile_sha256": "ce93dd7b9d59896ca6914ab8c4afd32d3fa98240b33a74f20a981cf909d8f25e"
    },
    {
      "id": "liquid-state-clear",
      "data_class": "selected_development_examples",
      "model_id": "liquid/d1-ultra",
      "display_name": "LiquidAI: d1-ultra",
      "returned_model": "liquid/d1-ultra-20261015",
      "provider": "Liquid",
      "mode": "state",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3157.0,
      "progress_pct": 100.0,
      "native_frames": 1255,
      "decisions": 314,
      "speedrun_score_native_frames": 1255,
      "nominal_game_seconds_to_completion": 20.916666666666668,
      "experiment_config": {
        "profile": "visible surfaces; movement plus A questions",
        "adapter": "movement_and_jump",
        "cadence": 4,
        "previous_screens": 0,
        "observation": "native current state and recent positions",
        "contract_sha256": "5b15378e4b1cf845d02cddb86455954d9e166a2393c2ed0ccedeadfa6937cd09",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "liquid-state-clear"
      },
      "input_audit": {
        "privileged_state_sent": true,
        "description": "state track intentionally receives current visible memory-derived state; independently replayed requests and outputs"
      },
      "trace_href": "traces/liquid-state-clear.json",
      "trace_sha256": "112a424917bb92b253f89d78eb96ac2d1242278d2174a2a04823fd03daae389a",
      "replay_href": "replays/liquid-state-clear.mp4",
      "replay_sha256": "d61d0b43b61666d0c55fb3f2a11e44e95b596aa1bc083de58d82ba965d02d4ff",
      "source_hashes": {
        "episode_sha256": "923582d608eee0eca83658b9c08a968d56048985ae1e7a1bf419e409abaf6ee4"
      },
      "controls": {
        "adapter": "movement_and_jump",
        "questions": [
          "jump",
          "movement"
        ],
        "trace_href": "traces/liquid-state-clear.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 1,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/liquid-state-clear.json",
      "profile_sha256": "74ed1ff9590ea1c33d1ba9fa0e0da58aaac849b9f9643240291271bdfd639a4d"
    },
    {
      "id": "pplx-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "perplexity/pplx-decider-v1.2",
      "display_name": "Perplexity: Decider V1.2",
      "returned_model": "perplexity/pplx-decider-v1.2-20261016",
      "provider": "Perplexity",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 2469.0,
      "progress_pct": 77.82761935277155,
      "native_frames": 1302,
      "decisions": 186,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "short edge semantics; explicit A and B choices",
        "adapter": "explicit_ab_joint12",
        "cadence": 7,
        "previous_screens": 1,
        "observation": "current full screen plus 1 past full screen; no RAM",
        "contract_sha256": "c25f2f228f875ed725c1f9044a70d18f08edad1612690b1b555b10f8c7f420fa",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "pplx-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 186,
        "description": "actual current/past screenshots and fixed pixel crops; declared rules/horizons and previous model controls ONLY",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/pplx-pixels-partial.json",
      "trace_sha256": "a3fc585f21f9f774d3ae0bb9b80af48832358158e873d237ab3e72d6b306d79d",
      "replay_href": "replays/pplx-pixels-partial.mp4",
      "replay_sha256": "fe44a6ba6f7da0325cd0d22ba2321cb85b15176a61d8bee46cbd75ca93472f14",
      "source_hashes": {
        "manifest_sha256": "22fca3a0aaf79205431590a3c40c62d92dfbdfecb5aa2b89caa974f824a16396",
        "requests_sha256": "421a36ebbc951e5dc355f200061140ee0291efe2a7ff4faec80191cb42cd04de",
        "responses_sha256": "77ec3925b542085646f6e4d75f391da5d71af0849176591358841b2708e4f739",
        "private_replay_sha256": "124ae679407fb24e38a5fbcca953ad4047f3ca739a51dfc4ae27c6fe6724cdb6",
        "input_audit_sha256": "483d1a8277d5af5ec13c1b50804f8ec0cb25ee09e7d94f85ede915453a4d270c"
      },
      "controls": {
        "adapter": "explicit_ab_joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/pplx-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/pplx-pixels-partial.json",
      "profile_sha256": "231d246ba671e176586b212ac4cbc35178c83c1973b497bcbf8fb26621f45fb7"
    },
    {
      "id": "jev-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "typesafe/jev-1.14",
      "display_name": "TypeSafe: Jev 1.14",
      "returned_model": "typesafe/jev-1.14-20261014",
      "provider": "TypeSafe",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1661.0,
      "progress_pct": 51.93848125600769,
      "native_frames": 1134,
      "decisions": 189,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "long-jump mechanics; four independent button questions",
        "adapter": "independent_buttons",
        "cadence": 6,
        "previous_screens": 1,
        "observation": "2 chronological full screens; no RAM",
        "contract_sha256": "27056f6bead483924a7ba096dfb72dd57148a4d389e3fbdddb45411eee74395b",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "jev-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 189,
        "description": "screenshots, declared rules/horizons and previous model controls only",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/jev-pixels-partial.json",
      "trace_sha256": "ab93c6e564f69416614ff71763b232280c7c8b4bc51341ec2e7416d014749909",
      "replay_href": "replays/jev-pixels-partial.mp4",
      "replay_sha256": "e5a690942ff1ffc78a936aada4dfe731ed4156a644913b653580cf9f57d7b828",
      "source_hashes": {
        "manifest_sha256": "f227b23f42c9900b547b829d407aff4319b68ae8ba68483ad852e5209920492d",
        "requests_sha256": "3aae5ed5534d33b2efd52b5b333e74a9f614dafb23b206f1487a7de90930bf1a",
        "responses_sha256": "e018715610a9f14c5e27529177fd7db9a0109f90ec6f3e9f77de2882973c9125",
        "private_replay_sha256": "7a7a14f2559bee48006f2a5328f214344b1d67aee1ce6a9e8a608e8926731625",
        "input_audit_sha256": "090b1d762940a5311dcc24c39aba45a118c9343ef02687944e00d9b8252d57f8"
      },
      "controls": {
        "adapter": "independent_buttons",
        "questions": [
          "horizontal",
          "jump",
          "run",
          "vertical"
        ],
        "trace_href": "traces/jev-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/jev-pixels-partial.json",
      "profile_sha256": "91400711a83623312598b9b60174fe068ba83cf43fd3dd0a87b95022ed936d18"
    },
    {
      "id": "mercury-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "inception/mercury-decide",
      "display_name": "Inception: Mercury Decide",
      "returned_model": "inception/mercury-decide-20260930",
      "provider": "Inception",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1797.0,
      "progress_pct": 56.29605895546299,
      "native_frames": 740,
      "decisions": 124,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "edgezoom; explicit A and B choices",
        "adapter": "explicit_ab_joint12",
        "cadence": 6,
        "previous_screens": 1,
        "observation": "current full screen, fixed left/middle crops, 1 past full screen; no RAM",
        "contract_sha256": "64d290fceaf2469f6037d490db6dddff02376c48a8d0d7c27fbb22bd1ccf6569",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "mercury-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 124,
        "description": "screenshots, declared rules/horizons and previous model controls only",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/mercury-pixels-partial.json",
      "trace_sha256": "1e06548f10b887b15f2278912e6d9538c4ff5153ec6e1de0c0f780a1af49a6aa",
      "replay_href": "replays/mercury-pixels-partial.mp4",
      "replay_sha256": "600ddef3f4f11b1c6db825642e9ca97c9ec71de87fec0ac3e4810b1f90d986c8",
      "source_hashes": {
        "manifest_sha256": "6144d74f8001126d45d7fa1479db133136dc744dcb1f295dc69787c8fc2c1edd",
        "requests_sha256": "d5c91fc570f58ad0db9c01b468a7bbbc92749aaff3091a19a2a7d6c1d9b3eb2e",
        "responses_sha256": "0407d3b7001e1ac51c05b00f861b481e5b08251cc600629be11b4357a57d6501",
        "private_replay_sha256": "d85302ebd02c8cbd8c779bcdf582b1e96e1b09131eb47cae23c0d34fee3a7ee1",
        "input_audit_sha256": "941cd6bf7b28644576bad3e8d1c815de601af2131e49a2db0ea23ecbe3bcdd36"
      },
      "controls": {
        "adapter": "explicit_ab_joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/mercury-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/mercury-pixels-partial.json",
      "profile_sha256": "dcf1d9a585735061ee13464d4831866d4f88139fffbed4959f324c500b1994ec"
    },
    {
      "id": "mammoth-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "baseten/mammoth-1",
      "display_name": "Baseten: Mammoth 1",
      "returned_model": "baseten/mammoth-1-20261021",
      "provider": "BaseTen",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1788.0,
      "progress_pct": 56.00768984299904,
      "native_frames": 866,
      "decisions": 145,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "edgezoom; explicit A and B choices",
        "adapter": "explicit_ab_joint12",
        "cadence": 6,
        "previous_screens": 1,
        "observation": "current full screen, fixed left/middle crops, 1 past full screen; no RAM",
        "contract_sha256": "64d290fceaf2469f6037d490db6dddff02376c48a8d0d7c27fbb22bd1ccf6569",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "mammoth-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 145,
        "description": "screenshots, declared rules/horizons and previous model controls only",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/mammoth-pixels-partial.json",
      "trace_sha256": "fe29c40485f38d73bb6a1441bf65950e04df867809e4b12c5b31c3fa812fbf8e",
      "replay_href": "replays/mammoth-pixels-partial.mp4",
      "replay_sha256": "3874b1fda33f539da566510e7f20182b6608e4e3cd7d394657ad5b75c768b191",
      "source_hashes": {
        "manifest_sha256": "e8bd2cd53528abc69ab8d3eb0d0a7aa0011e80101d68a78f5fbfa490142314e5",
        "requests_sha256": "10315b85ad9f0f6ea499fa10acffd05580bd6177567d093e3a2b77957fb137bf",
        "responses_sha256": "64791471a0869d630932eca27d27f4bfd1cb6071e070644b768ba8431a438752",
        "private_replay_sha256": "d909d35c82c34a951991a34550293c0470e9d39afe967f604a95596d77aa744d",
        "input_audit_sha256": "f15f630b0630bc3001add66bdb4109b06a12c3528d62aeb20ba3d5e877fce2df"
      },
      "controls": {
        "adapter": "explicit_ab_joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/mammoth-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/mammoth-pixels-partial.json",
      "profile_sha256": "28a034d1f76a7f6739c2a7c51b45010c81c542be2518dd0c6d17144510b643c5"
    },
    {
      "id": "mammoth-pixels-first-clear",
      "data_class": "selected_development_examples",
      "model_id": "baseten/mammoth-1",
      "display_name": "Baseten: Mammoth 1",
      "returned_model": "baseten/mammoth-1-20261021",
      "provider": "BaseTen",
      "mode": "image",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3157.0,
      "progress_pct": 100.0,
      "observed_progress_pct": 100.0,
      "native_frames": 1916,
      "decisions": 467,
      "speedrun_score_native_frames": 1916,
      "nominal_game_seconds_to_completion": 31.933333333333334,
      "experiment_config": {
        "profile": "memoryfinish240-v1",
        "adapter": "explicit_ab_joint12",
        "cadence": null,
        "initial_cadence_frames": 6,
        "cadence_mode": "controller-history-triggered native model-selected durations",
        "allowed_selected_durations": [
          2,
          4,
          6,
          8
        ],
        "previous_screens": 1,
        "observation": "current full screen, fixed left/middle crops and past full screen; declared prior model controls only",
        "contract_sha256": "429259115f9aaa9316e87287d8a8a7688bdaf3bf34898e232d296442e251530c",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "research_frame_budget": 18000,
        "research_decision_budget": 1600
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 467,
        "actual_pixel_images_audited": 1867,
        "native_replay_verified": true,
        "all_frames_verified": 1916,
        "duration_calls_audited": 255,
        "finish_rule_calls_audited": 7,
        "own_control_history_only": true,
        "independent_audit_pass": true
      },
      "trace_href": "traces/mammoth-pixels-first-clear.json",
      "trace_sha256": "58dadbda3e883f7892aefa5d723e4e8d50adfb2b77d7c4696fba4fcdcb9f3081",
      "profile_href": "profiles/mammoth-pixels-first-clear.json",
      "profile_sha256": "5c8ad5bf2566b8c828d574874b6c69e59093e0ad99841975fa3d5db7d4a7d209",
      "replay_href": "replays/mammoth-pixels-first-clear.mp4",
      "replay_sha256": "b9b1532ed8f97a75a544901849796d8b12db0de5e86f5aa0845cfb56324496b8",
      "replay_includes_initial_still": false,
      "proof_href": "proofs/mammoth-pixels-first-clear.json",
      "source_hashes": {
        "manifest_sha256": "fb31e92a607d29bf7a393065e9cb1fe330898f237b3e513cb93f96e1f2e377fd",
        "requests_sha256": "e3fb3994793fc6fa97858a292ffbed0b6eb3e633d2e2fd916c2b0bf0614f6642",
        "responses_sha256": "b403e00c43e5a9b1f4d7aa15c312192ead620e0c0590a6f1b3dbf94007c1e887",
        "private_replay_sha256": "39627b92d18c4586a4db95626aa863188cbde2c946ec084bdbebb0f57bc508e0",
        "profile_proof_sha256": "7e51d85c01c7b30451c6bbb8a114a2a29f82e76e20935a7910be684d46e85372",
        "input_native_audit_sha256": "a041a012f2dfa749a93f03ed2968c2e34e3b4fac5fb54f1612783f1a51f77123"
      },
      "controls": {
        "adapter": "native simultaneous buttons plus conditional native duration head",
        "trace_href": "traces/mammoth-pixels-first-clear.json"
      }
    }
  ],
  "protocol_sha256": "b50db77febc7519e1b7e5373d54bf69cf002c36a28e3b3bd297cb6191e7c0a54",
  "observations": [
    {
      "id": "state-c4-cloudflare--clef",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef",
      "display_name": "Cloudflare: Clef",
      "mode": "state",
      "level": "1-1",
      "returned_model": "cloudflare/clef",
      "provider": "PrimeIntellect",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1522.0,
      "progress_pct": 47.4847805190644,
      "observed_progress_pct": 47.4847805190644,
      "native_frames": 638,
      "decisions": 160,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-cloudflare--clef.json",
      "trace_sha256": "5a49c8c74e8b2b6460455feaa863f6eb5f90b08adcf6c71f099eab3a94ec5c5c",
      "profile_href": "profiles/state-c4-cloudflare--clef.json",
      "profile_sha256": "a6254712c15640da7c58d2d47f0256b5dbc0d4eff73932bc63fb40eee303244e",
      "replay_href": "replays/state-c4-cloudflare--clef.mp4",
      "replay_sha256": "0feda2bcc78e2760a1a947951f7875ba22482c07cf86a2f19334a6db55f453dd",
      "source_hashes": {
        "episode_sha256": "44417f19e9e77b0fcc72e29e6e512e6d04e65fc7d2b7f5ea67b8ecbdaf0c0e17",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "ed90a9c3daa74d40aa19a0e6346d308ca2aa4ebb727b41dea20f934e9b64f671"
      }
    },
    {
      "id": "state-c4-cloudflare--clef-flash",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef-flash",
      "display_name": "Cloudflare: Clef Flash",
      "mode": "state",
      "level": "1-1",
      "returned_model": "cloudflare/clef-flash",
      "provider": "PrimeIntellect",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 307.0,
      "progress_pct": 8.55495033643063,
      "observed_progress_pct": 8.55495033643063,
      "native_frames": 110,
      "decisions": 28,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-cloudflare--clef-flash.json",
      "trace_sha256": "1b05df1ca03979e2eaeee932f9175ef8bf25bba0265fe1050db284ff61215241",
      "profile_href": "profiles/state-c4-cloudflare--clef-flash.json",
      "profile_sha256": "052b6941ee8e7c41acc52cf7878315264bec4bdaaaed7ac4f74dcfdb61643ad7",
      "replay_href": "replays/state-c4-cloudflare--clef-flash.mp4",
      "replay_sha256": "182575c46405e5ea9ab2832759f514cbd57f84896ca606337cd12a949520c174",
      "source_hashes": {
        "episode_sha256": "3c47e092d1c30284fa967198cfd6e6a9ce8ac85adb35dac490fccbeb061275c9",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "b1a5a056b141f9ea25c79e4ab826bfe410407ffa2ad8c30ee2e0882eeef50f8b"
      }
    },
    {
      "id": "state-c4-cloudflare--clef-omni",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef-omni",
      "display_name": "Cloudflare: Clef Omni",
      "mode": "state",
      "level": "1-1",
      "returned_model": "cloudflare/clef-omni-20261009",
      "provider": "Cloudflare",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 451.0,
      "progress_pct": 13.168856135853893,
      "observed_progress_pct": 13.168856135853893,
      "native_frames": 1188,
      "decisions": 297,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-cloudflare--clef-omni.json",
      "trace_sha256": "79fdedeb3e85119d7c89b531b1901d3183b5c3b19f9bed241e4c95defc3962eb",
      "profile_href": "profiles/state-c4-cloudflare--clef-omni.json",
      "profile_sha256": "9da53e1879243e8a36e69f740eff90576c3faa73fe7b4bc6cb95240e15d972d9",
      "replay_href": "replays/state-c4-cloudflare--clef-omni.mp4",
      "replay_sha256": "7529b6361c57d60699d7cfabe81e803e8c3b903fa622e09bfa072d357a37d34f",
      "source_hashes": {
        "episode_sha256": "a95b8b618f37b2f5bfb28fa485b35d9034e05e5df03d2bc6854f934c80ae5d8c",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "35febc1f885ba7d4ec017264d5ec36426d5a937c231da25ebff6a99e7bb960e8"
      }
    },
    {
      "id": "state-c4-inception--mercury-decide",
      "data_class": "development_observation",
      "model_id": "inception/mercury-decide",
      "display_name": "Inception: Mercury Decide",
      "mode": "state",
      "level": "1-1",
      "returned_model": "inception/mercury-decide-20260930",
      "provider": "Inception",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 2019.0,
      "progress_pct": 63.409163729573855,
      "observed_progress_pct": 63.409163729573855,
      "native_frames": 840,
      "decisions": 210,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-inception--mercury-decide.json",
      "trace_sha256": "f9da99e068fe1a20c93c8ff7041aa6cda2d802c7a1f49c490621ce26810bb9bb",
      "profile_href": "profiles/state-c4-inception--mercury-decide.json",
      "profile_sha256": "37ff4f654409eeaa40f9540ffee999607a6a4b12edea0ff098a745ac04b94a83",
      "replay_href": "replays/state-c4-inception--mercury-decide.mp4",
      "replay_sha256": "f292aec244ef1c3cdee4e1aaadd42c842aa8cbc582b7bb85ed0debd8e57cbcb5",
      "source_hashes": {
        "episode_sha256": "6124e20c0f4485a5035efd4f19d66d935adf46a7574098680b2dd36966a6af4f",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "b2528355ef4707093777ad57556397c8aa4656c257260d4d990443119fdec01d"
      }
    },
    {
      "id": "state-c4-jaredpalmer--kev-4b",
      "data_class": "development_observation",
      "model_id": "jaredpalmer/kev-4b",
      "display_name": "Jared Palmer: Kev 4B",
      "mode": "state",
      "level": "1-1",
      "returned_model": null,
      "provider": null,
      "status": "api_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 40,
      "progress_pct": null,
      "observed_progress_pct": 0.0,
      "native_frames": 0,
      "decisions": 0,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-jaredpalmer--kev-4b.json",
      "trace_sha256": "b0f90913718a20bd3a64168128c6b46c3cb4cf427b71fc31de2b621d1a049321",
      "profile_href": "profiles/state-c4-jaredpalmer--kev-4b.json",
      "profile_sha256": "b9462dfcba655e1403fdf7a5eaef16882d46cb7eb144a37e4dfeddb1dc52995c",
      "replay_href": null,
      "replay_sha256": null,
      "source_hashes": {
        "episode_sha256": "8dccf56d4344c2a41aac46b9a34d5d53a99b70abd14d9121a2b02386afcbd9c7",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "192dc870b0fc217a3c9521b2d0d896c68d3b6759765aef0286d67084af7042e2"
      },
      "replay_unavailable_reason": "no native gameplay frames: the api attempt ended before any action"
    },
    {
      "id": "state-c4-microsoft--microsoft-decision-1",
      "data_class": "development_observation",
      "model_id": "microsoft/microsoft-decision-1",
      "display_name": "Microsoft: Microsoft-Decision-1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "microsoft/microsoft-decision-1-20261009",
      "provider": "Azure",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1421.0,
      "progress_pct": 44.24863825696892,
      "observed_progress_pct": 44.24863825696892,
      "native_frames": 632,
      "decisions": 158,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-microsoft--microsoft-decision-1.json",
      "trace_sha256": "ca9e679bc4bdf0aaf9fa413598768eadaa265e2d24e5d5bbb3a4a83a522cb317",
      "profile_href": "profiles/state-c4-microsoft--microsoft-decision-1.json",
      "profile_sha256": "0af912cc8596bff46c924be4ccc627d3fa4e61fb79d58b5e8e28d693e4cc6994",
      "replay_href": "replays/state-c4-microsoft--microsoft-decision-1.mp4",
      "replay_sha256": "dd791bef04d6cbf19183c7b31ec40b829b6867e8c82b43c1eba4935590a57847",
      "source_hashes": {
        "episode_sha256": "feaa110e53327baef1b1e09393d066387d83d9d1c64b7c70cef8d0cef0a442ca",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "5ac4945a062fde78f54ad97ebfd461af88b52eb852c54efc40d6adbc1d71a919"
      }
    },
    {
      "id": "state-c4-nace-ai--drex-v1-5",
      "data_class": "development_observation",
      "model_id": "nace-ai/drex-v1.5",
      "display_name": "Nace.AI: Drex v1.5",
      "mode": "state",
      "level": "1-1",
      "returned_model": "nace-ai/drex-v1.5-20261009",
      "provider": "DeepInfra",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 596.0,
      "progress_pct": 17.81480294777315,
      "observed_progress_pct": 17.81480294777315,
      "native_frames": 840,
      "decisions": 210,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-nace-ai--drex-v1-5.json",
      "trace_sha256": "1856a7c0ca2c21f4fd3e3ce9c0ae765217ee61138705379e5af0b4b6dd383b29",
      "profile_href": "profiles/state-c4-nace-ai--drex-v1-5.json",
      "profile_sha256": "2764143a2334246de9a969856fd85b49e2d7dd8ebfaec8cfd7a60113850165a2",
      "replay_href": "replays/state-c4-nace-ai--drex-v1-5.mp4",
      "replay_sha256": "7b647a8b5b246021dd7aa2c8b6038fe1cb684fd470b9bfcc64b6fd49c0a82796",
      "source_hashes": {
        "episode_sha256": "51af315a2f1295db170543d9b18efab63a9838b564ea2c7b19464c70319c5a67",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "b4d1c2863dfe0c35784b866e6482ce313673e11595f263c913eeb1b6df8987a1"
      }
    },
    {
      "id": "state-c4-nace-ai--drex-v2-0",
      "data_class": "development_observation",
      "model_id": "nace-ai/drex-v2.0",
      "display_name": "Nace.AI: Drex v2.0",
      "mode": "state",
      "level": "1-1",
      "returned_model": "nace-ai/drex-v2.0-20261009",
      "provider": "DeepInfra",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 594.0,
      "progress_pct": 17.75072092278116,
      "observed_progress_pct": 17.75072092278116,
      "native_frames": 1932,
      "decisions": 483,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-nace-ai--drex-v2-0.json",
      "trace_sha256": "e4f3e0e866fc09adb6b83ec33fb54786d7935a834b12a047b3f1d5c68c7f7ae4",
      "profile_href": "profiles/state-c4-nace-ai--drex-v2-0.json",
      "profile_sha256": "0c7caf7cbaeafc9397f72607fd21f674138b8d3b3fd88669b4a2fce1308e3a55",
      "replay_href": "replays/state-c4-nace-ai--drex-v2-0.mp4",
      "replay_sha256": "7102dda237bed2052ed730bf61567599687dbc27a7e2f6c69374abfa3ebcd2fa",
      "source_hashes": {
        "episode_sha256": "b72c05484d921ff7bb76c7606b2485ad917a9fcbfdf998b2e1305e6eb6d10086",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "f800912ec89dacb4a3aedefa2b56fe90dba22f1eeb62c246084591417cb44790"
      }
    },
    {
      "id": "state-c4-perplexity--pplx-decider-flash-v1-2",
      "data_class": "development_observation",
      "model_id": "perplexity/pplx-decider-flash-v1.2",
      "display_name": "Perplexity: Decider Flash V1.2",
      "mode": "state",
      "level": "1-1",
      "returned_model": "perplexity/pplx-decider-flash-v1.2-20261016",
      "provider": "Perplexity",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 312.0,
      "progress_pct": 8.715155398910605,
      "observed_progress_pct": 8.715155398910605,
      "native_frames": 106,
      "decisions": 27,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-perplexity--pplx-decider-flash-v1-2.json",
      "trace_sha256": "30ebc59f6d37348046620bd59c293e94076a7d6e0c09ea4fe576c3f8c788260c",
      "profile_href": "profiles/state-c4-perplexity--pplx-decider-flash-v1-2.json",
      "profile_sha256": "0035cc6eace31e6edc9875ed49142f48c859338384f386a019028f02d8bd7084",
      "replay_href": "replays/state-c4-perplexity--pplx-decider-flash-v1-2.mp4",
      "replay_sha256": "6fed06d0f41e468fcf9adbd855517291e8f97de869d07754cad4e6965e1a395e",
      "source_hashes": {
        "episode_sha256": "1097eca1d98ee816bc523cdb5d80e2118e576f258ee688244c787ba4e880f05d",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "28bc23a37b2c648079a76ac189849bd1999366e87e3df1365acbe6f2c03ebf3e"
      }
    },
    {
      "id": "state-c4-respan--span-02",
      "data_class": "development_observation",
      "model_id": "respan/span-02",
      "display_name": "Respan: Span-02",
      "mode": "state",
      "level": "1-1",
      "returned_model": "respan/span-02-20261015",
      "provider": "Respan",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1786.0,
      "progress_pct": 55.94360781800705,
      "observed_progress_pct": 55.94360781800705,
      "native_frames": 776,
      "decisions": 194,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-respan--span-02.json",
      "trace_sha256": "1da3204711b9fa613c8b97e87659e2cd2736f19d78bf7baa7e715d294d1a2917",
      "profile_href": "profiles/state-c4-respan--span-02.json",
      "profile_sha256": "0a221aa5e696d593aebd67a2446b5c72d7f85c7fef6993b560ce51ceed65adf0",
      "replay_href": "replays/state-c4-respan--span-02.mp4",
      "replay_sha256": "b19a9ee4150a6e4ac6bc37376ffada49079f9586ede4819c407bf1c184f21182",
      "source_hashes": {
        "episode_sha256": "10939a887b76060ca88eb927e1d70ca7546b79cdb13fc4f69a2d4488c2f00c07",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "b92dd09a60e87765d6853100869f72c2916338cf54887178c06c45c28ce10987"
      }
    },
    {
      "id": "state-c4-scaledown--classify-1",
      "data_class": "development_observation",
      "model_id": "scaledown/classify-1",
      "display_name": "ScaleDown: Classify-1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "scaledown/classify-1-20260928",
      "provider": "ScaleDown",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 40,
      "progress_pct": 0.0,
      "observed_progress_pct": 0.0,
      "native_frames": 604,
      "decisions": 151,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-scaledown--classify-1.json",
      "trace_sha256": "38635d43931cf2e926c51e89ffd15814738830bbb58be69ddc5947d351375a0f",
      "profile_href": "profiles/state-c4-scaledown--classify-1.json",
      "profile_sha256": "23ae0ce182266b81847a041add76bfce2d4a54c9d81422a97b584828839b1d16",
      "replay_href": "replays/state-c4-scaledown--classify-1.mp4",
      "replay_sha256": "62c55f027d19e2eb1d6aa3ff5e7eb80e053d55eced33ae05510d4ed54dcb84cc",
      "source_hashes": {
        "episode_sha256": "f706a76dcb4ae1784f4651d346c05449a48928037c67fc96a5e521a28c8ef01e",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "b0278817cd43f5f9e49ec21a848a1bf4a4d3ed3227348f267325a02d073f169a"
      }
    },
    {
      "id": "state-c4-typesafe--jev-1-13",
      "data_class": "development_observation",
      "model_id": "typesafe/jev-1.13",
      "display_name": "TypeSafe: Jev 1.13",
      "mode": "state",
      "level": "1-1",
      "returned_model": "typesafe/jev-1.13-20260917",
      "provider": "TypeSafe",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 723.0,
      "progress_pct": 21.884011534764497,
      "observed_progress_pct": 21.884011534764497,
      "native_frames": 904,
      "decisions": 226,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-typesafe--jev-1-13.json",
      "trace_sha256": "1c2f2f228d20a25223239a1647c99f39117c8e4d666f5f7c1bee765c4b5bc500",
      "profile_href": "profiles/state-c4-typesafe--jev-1-13.json",
      "profile_sha256": "b7cc8102297007adc52f1f5a3ed0f5f58670f0bde3452f1debaa915292fb3d6d",
      "replay_href": "replays/state-c4-typesafe--jev-1-13.mp4",
      "replay_sha256": "27e03c0b05f5a9e31bc21a15f23dd1225c42127ae79942adda76b1fd132991b2",
      "source_hashes": {
        "episode_sha256": "dc708992ce95cdb09a5a31c85f2f2732842bb0205f72020ebd3ec5edb99fad19",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "2a94f715716cb7018b2de7ff3d61706e0166ee3f231a5352c11faa7434c7b434"
      }
    },
    {
      "id": "state-c4-typesafe--jev-1-14",
      "data_class": "development_observation",
      "model_id": "typesafe/jev-1.14",
      "display_name": "TypeSafe: Jev 1.14",
      "mode": "state",
      "level": "1-1",
      "returned_model": "typesafe/jev-1.14-20261014",
      "provider": "TypeSafe",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 596.0,
      "progress_pct": 17.81480294777315,
      "observed_progress_pct": 17.81480294777315,
      "native_frames": 884,
      "decisions": 221,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-typesafe--jev-1-14.json",
      "trace_sha256": "42720bb8886597ffba25d5c9cb4f9712ae4e4497ebe4a359591ea8dd99b7e2d7",
      "profile_href": "profiles/state-c4-typesafe--jev-1-14.json",
      "profile_sha256": "d35b723ba847fdf0de774cb3890cb455c2ce37ce29bfa1959c08f44d5ed1f5cf",
      "replay_href": "replays/state-c4-typesafe--jev-1-14.mp4",
      "replay_sha256": "9ea3c9b2c888ad4faebccfebdc2bf20e4cc16d4bbc913cc96d805401a3bf7a1d",
      "source_hashes": {
        "episode_sha256": "9f0a5dddc3e2c164e6d736642f3a6158a433632145a4be5b95072dfbe200a742",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "5c0116cc4ed5ecb521aed7334b3faa36da477d8a60dc39847c2441bccd9d57e6"
      }
    },
    {
      "id": "state-c4-upstage--solar-decide-flash",
      "data_class": "development_observation",
      "model_id": "upstage/solar-decide-flash",
      "display_name": "Upstage: Solar Decide Flash",
      "mode": "state",
      "level": "1-1",
      "returned_model": "upstage/solar-decide-flash-20261008",
      "provider": "Upstage",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 40,
      "progress_pct": 0.0,
      "observed_progress_pct": 0.0,
      "native_frames": 604,
      "decisions": 151,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-upstage--solar-decide-flash.json",
      "trace_sha256": "2dfd75afda7a584d3deba3044b2eee4050cee5f21db076e4c487b8f27bd01db2",
      "profile_href": "profiles/state-c4-upstage--solar-decide-flash.json",
      "profile_sha256": "5cb688968a5589a308c84614d5e996a33b8cedbf4e1e7cdcf9dc7a85a4efc3b0",
      "replay_href": "replays/state-c4-upstage--solar-decide-flash.mp4",
      "replay_sha256": "2f5d68227911f7ba902ebd50af98eb18c8fec58897027e51e056e5eaca287e70",
      "source_hashes": {
        "episode_sha256": "b8f44a4108ab05cf7d98907e206b394483659dcf132405885bf08a74046bf692",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "187cbd88130a9e50fcf3ff2774761fe2c7f62ce12719fcbcc6cb9277d7fd85c0"
      }
    },
    {
      "id": "state-c4-baseten--elephant-1",
      "data_class": "development_observation",
      "model_id": "baseten/elephant-1",
      "display_name": "Baseten: Elephant 1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "baseten/elephant-1-20261021",
      "provider": "BaseTen",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 313.0,
      "progress_pct": 8.7471964114066,
      "observed_progress_pct": 8.7471964114066,
      "native_frames": 122,
      "decisions": 31,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-baseten--elephant-1.json",
      "trace_sha256": "0464ade2ebe92f293b843559c7107f80f31e9a20118987a26ea712f5bd1515c4",
      "profile_href": "profiles/state-c4-baseten--elephant-1.json",
      "profile_sha256": "6cdc423a9e72eb783114c4ab656813d21ceb4ab640c5d90fba208da9f56ca0eb",
      "replay_href": "replays/state-c4-baseten--elephant-1.mp4",
      "replay_sha256": "8f057bd61c2bd51a13f0959586ff54f152caada39ff9c2b0910fa2080578cf08",
      "source_hashes": {
        "episode_sha256": "84bea44c52d753959b282d78c33451a7b7a18521d0f6d6537801598eb65895c0",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "bb8235a91a5f5ba198f3200f7231e47fc1a55db9b0c39ff9eaef739a1fdfebd8"
      }
    },
    {
      "id": "state-c4-baseten--joey-1",
      "data_class": "development_observation",
      "model_id": "baseten/joey-1",
      "display_name": "Baseten: Joey 1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "baseten/joey-1-20261029",
      "provider": "BaseTen",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 312.0,
      "progress_pct": 8.715155398910605,
      "observed_progress_pct": 8.715155398910605,
      "native_frames": 106,
      "decisions": 27,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-baseten--joey-1.json",
      "trace_sha256": "a1544f4fc9101f0d9426e3b15f8bd8f21da57646f07f898c988c6aeb466be0c4",
      "profile_href": "profiles/state-c4-baseten--joey-1.json",
      "profile_sha256": "1087de48eaa3ce3c9407438475598fcf3f9dcfb3d31d0495c5895cce151c6b89",
      "replay_href": "replays/state-c4-perplexity--pplx-decider-flash-v1-2.mp4",
      "replay_sha256": "6fed06d0f41e468fcf9adbd855517291e8f97de869d07754cad4e6965e1a395e",
      "source_hashes": {
        "episode_sha256": "c346b0523909da5ffd7b131038f04d2f9607c1cef25e87a0b634187f2a8898ef",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "4a77c8b3149d7fff9f2625339858928a83e21afbf2ec5f00c6e8dee73e49ed3b"
      }
    },
    {
      "id": "state-c4-baseten--mammoth-1",
      "data_class": "development_observation",
      "model_id": "baseten/mammoth-1",
      "display_name": "Baseten: Mammoth 1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "baseten/mammoth-1-20261021",
      "provider": "BaseTen",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 596.0,
      "progress_pct": 17.81480294777315,
      "observed_progress_pct": 17.81480294777315,
      "native_frames": 804,
      "decisions": 201,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-baseten--mammoth-1.json",
      "trace_sha256": "738a78e9b01466c9eede0202872be6443e123eb48ee0a433f6ae5b1b24df0efb",
      "profile_href": "profiles/state-c4-baseten--mammoth-1.json",
      "profile_sha256": "c3b217d0884fed1c0e90592249ff82053af51051d1acf8f2a3a48a7f334d1ae3",
      "replay_href": "replays/state-c4-baseten--mammoth-1.mp4",
      "replay_sha256": "4b63e16cf32179e879a84ad9350e28eb87cb171f9f51241f95f5c2f3fc5d99b8",
      "source_hashes": {
        "episode_sha256": "ad1f40f7002a2dc2c37edad647a442c22719431ba91c7a310538862c86f34621",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "fa95f6cfa13398333c912663de7ccb6ffb6a436004046da0c88bfc0df3975e9d"
      }
    },
    {
      "id": "state-c4-liquid--d1-ultra",
      "data_class": "development_observation",
      "model_id": "liquid/d1-ultra",
      "display_name": "LiquidAI: d1-ultra",
      "mode": "state",
      "level": "1-1",
      "returned_model": "liquid/d1-ultra-20261015",
      "provider": "Liquid",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 679.0,
      "progress_pct": 20.474206984940725,
      "observed_progress_pct": 20.474206984940725,
      "native_frames": 282,
      "decisions": 71,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-liquid--d1-ultra.json",
      "trace_sha256": "bb648446fada8825c7595c66e665904e8aa966edb50dcfa3af4dee201aa32ce3",
      "profile_href": "profiles/state-c4-liquid--d1-ultra.json",
      "profile_sha256": "446d9d62b14bc7ef0a1fbb98f299dc11b81319eda57b73a59c6613d9bd856e48",
      "replay_href": "replays/state-c4-liquid--d1-ultra.mp4",
      "replay_sha256": "a93c20ad0dbfe030de2be347f5c65cd0767eac80370ff69032d1baf455fb39a0",
      "source_hashes": {
        "episode_sha256": "23099283b5794a7d0ec1e588f6ddbda8b4de94746542f49f55ac39783e5f8b5e",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "fe253e29ca7ed8a28ac5044cd88c7ee9928826a68c52aa9f325ef92ea8b7cc8d"
      }
    },
    {
      "id": "state-c4-liquid--d1-1",
      "data_class": "development_observation",
      "model_id": "liquid/d1.1",
      "display_name": "LiquidAI: d1.1",
      "mode": "state",
      "level": "1-1",
      "returned_model": "liquid/d1.1-20261027",
      "provider": "Liquid",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 312.0,
      "progress_pct": 8.715155398910605,
      "observed_progress_pct": 8.715155398910605,
      "native_frames": 106,
      "decisions": 27,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-liquid--d1-1.json",
      "trace_sha256": "58fb00c2c336ebd746834acbbe335d42f2e41cfc079405e03134c6347cf88bdc",
      "profile_href": "profiles/state-c4-liquid--d1-1.json",
      "profile_sha256": "365413c0ba191c93aeb1d3b5a77ce8d69fc6544c01ecbf61d4603bcd75caa6d7",
      "replay_href": "replays/state-c4-perplexity--pplx-decider-flash-v1-2.mp4",
      "replay_sha256": "6fed06d0f41e468fcf9adbd855517291e8f97de869d07754cad4e6965e1a395e",
      "source_hashes": {
        "episode_sha256": "5e85b7cefeba1bcda94cbd4c59e900c172a0447ed8ec7dc41d2a282e156172ce",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "f5952453f0e6919dae2ee00fffdeb3c09b3f93c45d11be4836870cae836e32cf"
      }
    },
    {
      "id": "state-c4-nace-ai--drex-max",
      "data_class": "development_observation",
      "model_id": "nace-ai/drex-max",
      "display_name": "Nace.AI: Drex Max",
      "mode": "state",
      "level": "1-1",
      "returned_model": "nace-ai/drex-max-20261016",
      "provider": "DeepInfra",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 670.0,
      "progress_pct": 20.18583787247677,
      "observed_progress_pct": 20.18583787247677,
      "native_frames": 244,
      "decisions": 61,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-nace-ai--drex-max.json",
      "trace_sha256": "726037d8ef935db5f2c5a85b8cfbb91f0ee238af6693af9f9d5a32e5a9133700",
      "profile_href": "profiles/state-c4-nace-ai--drex-max.json",
      "profile_sha256": "31d640acf95eee4d2aad0c9bf6dc65f2e49f31f36a3eb004f1b9c51a54dd3eed",
      "replay_href": "replays/state-c4-nace-ai--drex-max.mp4",
      "replay_sha256": "7051f6fd0daea9a82e00dd4c07719d641a60d6611a529e386e043e83617b25b8",
      "source_hashes": {
        "episode_sha256": "18616e77a0397cbd09a0289d78ed354bfdbe159c33592e7df138869a65695971",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "587e0b2676d9fbee78bd333babeca7b523672293d64988b68f47b5bb9c89d06e"
      }
    },
    {
      "id": "state-c4-openai--gpt-6-luna-decisions",
      "data_class": "development_observation",
      "model_id": "openai/gpt-6-luna-decisions",
      "display_name": "OpenAI: GPT-6 Luna Decisions",
      "mode": "state",
      "level": "1-1",
      "returned_model": "openai/gpt-6-luna-decisions-20261006",
      "provider": "OpenAI",
      "status": "api_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 596.0,
      "progress_pct": null,
      "observed_progress_pct": 17.81480294777315,
      "native_frames": 284,
      "decisions": 71,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "shared-phase-c4-development",
        "cadence": 4,
        "history": 0,
        "contract_sha256": "8b6c487581b9bf4e0797833729bc0edd94137cc260ecbabd6cd04e2b192662d8",
        "matched_cohort": false,
        "selected_attempt": false,
        "maximum_native_frames": 6000,
        "development_stall_abort_after_frames": 600
      },
      "input_audit": {
        "native_replay_verified": true,
        "exact_request_audit_complete": true,
        "privileged_state_sent": true
      },
      "trace_href": "traces/state-c4-openai--gpt-6-luna-decisions.json",
      "trace_sha256": "726b84b3758b7c7455a6d70d8b2c7f3b2ad5762add7ef0f0d7156166d4ca8be0",
      "profile_href": "profiles/state-c4-openai--gpt-6-luna-decisions.json",
      "profile_sha256": "56a9fc9e24a609f36d382835fba94b9afd118035f302e150f6f9e57cdd88d51e",
      "replay_href": "replays/state-c4-openai--gpt-6-luna-decisions.mp4",
      "replay_sha256": "3d521a5044438bdf92d4f6306ad3aec2a5350c3adc3aa83ea0908f2363683031",
      "source_hashes": {
        "episode_sha256": "388987085e2a74c3e58684f28392c63b712d3e6132e7a4be3400d3198a41260b",
        "observer_source_sha256": "e9a16a88c9fbffbe62d62b7ce4f17714a764e5430528014ccb39cc9e23a2529c",
        "audit_sha256": "cd01c692af00d7fde1c364c6752bce5af554a49e5a8f80fcd36a74d4ddfd5ea4"
      }
    },
    {
      "id": "pplx-state-clear",
      "data_class": "selected_development_examples",
      "model_id": "perplexity/pplx-decider-v1.2",
      "display_name": "Perplexity: Decider V1.2",
      "returned_model": "perplexity/pplx-decider-v1.2-20261016",
      "provider": "Perplexity",
      "mode": "state",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3155.0,
      "progress_pct": 100.0,
      "native_frames": 1208,
      "decisions": 302,
      "speedrun_score_native_frames": 1208,
      "nominal_game_seconds_to_completion": 20.133333333333333,
      "experiment_config": {
        "profile": "phase guidance; compact visible state",
        "adapter": "joint12",
        "cadence": 4,
        "previous_screens": 0,
        "observation": "native current state and recent positions",
        "contract_sha256": "905208dc2a10e81fc31923f84ca614968e21979de9141a8f2409c238a4e63a3c",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "pplx-state-clear"
      },
      "input_audit": {
        "privileged_state_sent": true,
        "description": "state track intentionally receives current visible memory-derived state; independently replayed requests and outputs"
      },
      "trace_href": "traces/pplx-state-clear.json",
      "trace_sha256": "f2b470ca6ce60e07f60a69a36d1a965db44bbde75abc81da51038a6f5be02891",
      "replay_href": "replays/pplx-state-clear.mp4",
      "replay_sha256": "06949c052e4823c82b4736de0e67f074d7b532237dffbc2fc6f2fb723b5844b6",
      "source_hashes": {
        "episode_sha256": "e2e503b1067ec1e485c5e82cc88581868341a5544d9ffc8f794a92e12c0bbcdb"
      },
      "controls": {
        "adapter": "joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/pplx-state-clear.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 1,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/pplx-state-clear.json",
      "profile_sha256": "05b21e926579c535547784e784dc1608ce19cd61ecf677567eaddc3ce9c2a8ec",
      "cohort_note": "Same input profile and cadence; selected prior clear used 2400 decision cap, cleared after302 under common1500 cap; post-hoc inclusion, not prespecified"
    },
    {
      "id": "pixels-best-baseten--elephant-1",
      "data_class": "development_observation",
      "model_id": "baseten/elephant-1",
      "display_name": "Baseten: Elephant 1",
      "mode": "image",
      "level": "1-1",
      "returned_model": "baseten/elephant-1-20261021",
      "provider": "BaseTen",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1956.0,
      "progress_pct": 61.39057994232618,
      "observed_progress_pct": 61.39057994232618,
      "native_frames": 808,
      "decisions": 50,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "macros",
        "cadence": 8,
        "history": 1,
        "contract_sha256": "233803053de136b284dd443a98a243c1a50dbd1cfda5f466b8af934e71b7aeeb",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 50
      },
      "trace_href": "traces/pixels-best-baseten--elephant-1.json",
      "trace_sha256": "e631fd8a9be1289c0275cfe7d02cb3ae1134f923e57855c3915e386ace6bf167",
      "profile_href": "profiles/pixels-best-baseten--elephant-1.json",
      "profile_sha256": "5f2ad5f14325ad9d80766db5a339b35e667ae4522a361df7b3dc81d51dcd8684",
      "replay_href": "replays/pixels-best-baseten--elephant-1.mp4",
      "replay_sha256": "8926249fd1f65a7d5fe1b847e4bf6beecbda0a946fb91520820a9f85fb1f67ad",
      "source_hashes": {
        "requests_sha256": "51e2a06c0562cc15c005a941e3face4bfa274b3f9a90e37f1aeefbcacaab7f19",
        "responses_sha256": "f875111769d7b234c944b1e617a3e7289bd4dff929b09f33c10098e1d5b72389",
        "private_replay_sha256": "6eb25179d96296a27a7c60cb6daf6b8931c061077e5ee7fbab3ca4579d76800c",
        "audit_sha256": "9d27c14fff30f20f8c2ff86720f0bd7a3fa856a55112adf665ea9c42c2a82551"
      }
    },
    {
      "id": "pixels-best-baseten--joey-1",
      "data_class": "development_observation",
      "model_id": "baseten/joey-1",
      "display_name": "Baseten: Joey 1",
      "mode": "image",
      "level": "1-1",
      "returned_model": "baseten/joey-1-20261029",
      "provider": "BaseTen",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 693.0,
      "progress_pct": 20.922781159884654,
      "observed_progress_pct": 20.922781159884654,
      "native_frames": 312,
      "decisions": 13,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "pulse24",
        "cadence": 8,
        "history": 2,
        "contract_sha256": "30663b6a3c7faac1237201b209e3fffc4a2539d8215f57a452b5c153c523f070",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 13
      },
      "trace_href": "traces/pixels-best-baseten--joey-1.json",
      "trace_sha256": "f61cbd01a486871fcec7490b5d3fbf9dd3cf6ed1980ccb8989b1722f0d43feb0",
      "profile_href": "profiles/pixels-best-baseten--joey-1.json",
      "profile_sha256": "6ab07db093ac3519db99a49a50387244e710b59296f82653e76190610eed26ea",
      "replay_href": "replays/pixels-best-baseten--joey-1.mp4",
      "replay_sha256": "a798984c720b4e4b5f7b8389edc6f45b15c4a6d928a26a980bf1355e79194ee1",
      "source_hashes": {
        "requests_sha256": "1785598002d428b6622ecd1d7d92e8084a3874b74c71a9354a5034cb1b7299f3",
        "responses_sha256": "9a70e44c43260cabddee2624475be0cc4aa852a453c5dfca0992655dbcfd35e3",
        "private_replay_sha256": "70694c913b704e4627178f07c8d07c315bfcd4ac5d83852f7ba5d17d9bfd1bfd",
        "audit_sha256": "d476d00ed52ff6fbfec16d4fc5648db5583f984598c4858fb726953c1020311f"
      }
    },
    {
      "id": "pixels-best-baseten--mammoth-1",
      "data_class": "development_observation",
      "model_id": "baseten/mammoth-1",
      "display_name": "Baseten: Mammoth 1",
      "mode": "image",
      "level": "1-1",
      "returned_model": "baseten/mammoth-1-20261021",
      "provider": "BaseTen",
      "status": "development_stagnation_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 2594.0,
      "progress_pct": 81.83274591477091,
      "observed_progress_pct": 81.83274591477091,
      "native_frames": 1920,
      "decisions": 320,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "runup",
        "cadence": 6,
        "history": 2,
        "contract_sha256": "7b13971a0e18058633938b815d5043d0114057d793beba8a437903a27149d0e4",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 320
      },
      "trace_href": "traces/pixels-best-baseten--mammoth-1.json",
      "trace_sha256": "1cb87de44c1b80dc143949e7ed814920ebde3c5137a6cacaca3b40d0935cc140",
      "profile_href": "profiles/pixels-best-baseten--mammoth-1.json",
      "profile_sha256": "d295311acc09af5aa8fab5f33d73d8a43083988e33a18e246a83b9884c645ed8",
      "replay_href": "replays/pixels-best-baseten--mammoth-1.mp4",
      "replay_sha256": "d231a80b938dbd8c2e202e103306d21a69941282a54120eb6c8c7e42893b2137",
      "source_hashes": {
        "requests_sha256": "4d4a744ea4fea04d46b2603067397897e230b04046167386973ddc13c01f5d09",
        "responses_sha256": "a65941302218764a47f95c47cfb0926a6eb05f6ec649884184a6ea0571bdff52",
        "private_replay_sha256": "35d9cd84232ac2ced177e4e0d0769a275af09172c1c6c46079dbe6f35bb0d122",
        "audit_sha256": "dd0e404a2b78ce1e09ac4f8cce30d423ecc739416ca15f05c3a95aa3949e5e6c"
      }
    },
    {
      "id": "pixels-best-cloudflare--clef",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef",
      "display_name": "Cloudflare: Clef",
      "mode": "image",
      "level": "1-1",
      "returned_model": "cloudflare/clef",
      "provider": "Cloudflare",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1125.0,
      "progress_pct": 34.764498558154436,
      "observed_progress_pct": 34.764498558154436,
      "native_frames": 514,
      "decisions": 21,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "smartmacros",
        "cadence": 8,
        "history": 1,
        "contract_sha256": "a064eacd76eb6dc3c53c015b14a55e4c2275aed6c06944d3e710b2ad6bc96ab3",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 21
      },
      "trace_href": "traces/pixels-best-cloudflare--clef.json",
      "trace_sha256": "605b198ce395653f286359d07c468e30630388b883074347237986e1273dfa48",
      "profile_href": "profiles/pixels-best-cloudflare--clef.json",
      "profile_sha256": "ddb1bb4112fccab4d4f5ffc820255fe51ef6377a58336197faba6e9023c02603",
      "replay_href": "replays/pixels-best-cloudflare--clef.mp4",
      "replay_sha256": "d7c0ff99057f77d004dfd62c21c4a68dd03f2d104aa560748c75a4f0d0ac094e",
      "source_hashes": {
        "requests_sha256": "5f6506d3f325625483ac328ddac266849054db1a94f66ee93621f0fd665b5fb8",
        "responses_sha256": "7e0f4988c99db984ed164154583167c8d32994f9c0fc6862a4d6d390eeb4a12d",
        "private_replay_sha256": "52d62dfd9e37f7e583eda4716947594771c45e28369434315cc60952a4281b46",
        "audit_sha256": "663cf995a502245cb2a98061243197673f34a5b0668d595bddc0e0a4a255d57a"
      }
    },
    {
      "id": "pixels-best-cloudflare--clef-flash",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef-flash",
      "display_name": "Cloudflare: Clef Flash",
      "mode": "image",
      "level": "1-1",
      "returned_model": "cloudflare/clef-flash",
      "provider": "Cloudflare",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1125.0,
      "progress_pct": 34.764498558154436,
      "observed_progress_pct": 34.764498558154436,
      "native_frames": 511,
      "decisions": 17,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "smartmacros",
        "cadence": 8,
        "history": 1,
        "contract_sha256": "a064eacd76eb6dc3c53c015b14a55e4c2275aed6c06944d3e710b2ad6bc96ab3",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 17
      },
      "trace_href": "traces/pixels-best-cloudflare--clef-flash.json",
      "trace_sha256": "de851a72a001d1b19d115d371372cf9501bde13ca58fc796597b6d2b31352d80",
      "profile_href": "profiles/pixels-best-cloudflare--clef-flash.json",
      "profile_sha256": "69f90b0bf48577324bf0c4f1482df6b58a747ac2187d154df0efda20738e0de9",
      "replay_href": "replays/pixels-best-cloudflare--clef-flash.mp4",
      "replay_sha256": "3059f337550eff2c0189e9bb1ca8c1b5ee6cc1515321a7eff05415177b9ddfb9",
      "source_hashes": {
        "requests_sha256": "365d56cd3229ea227d1af67de517ac820e02534152b726bde84f55ee1a11250e",
        "responses_sha256": "8e544a3b2597773027081fb8f0f23820105c4f1a80ab447ef98119699c78855c",
        "private_replay_sha256": "7acbaf90eb7b945e7894814bbfc30aa19f035139f833e8a3453fd4483096ca16",
        "audit_sha256": "d10cc5f5e567403ea1427c68e0b2658fa2370ec93c6537ec7ad07242feda389b"
      }
    },
    {
      "id": "pixels-best-cloudflare--clef-omni",
      "data_class": "development_observation",
      "model_id": "cloudflare/clef-omni",
      "display_name": "Cloudflare: Clef Omni",
      "mode": "image",
      "level": "1-1",
      "returned_model": "cloudflare/clef-omni-20261009",
      "provider": "Cloudflare",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 672.0,
      "progress_pct": 20.24991989746876,
      "observed_progress_pct": 20.24991989746876,
      "native_frames": 244,
      "decisions": 12,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "pulse24",
        "cadence": 8,
        "history": 2,
        "contract_sha256": "30663b6a3c7faac1237201b209e3fffc4a2539d8215f57a452b5c153c523f070",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 12
      },
      "trace_href": "traces/pixels-best-cloudflare--clef-omni.json",
      "trace_sha256": "cb02b68730cc2189fdbde410933168aa855c7a49ce34a6393343462d154c69aa",
      "profile_href": "profiles/pixels-best-cloudflare--clef-omni.json",
      "profile_sha256": "0fd3e44ecbf22b3962ee92a0e84acfbfb4165203b887aeee74381da9ccd9c54e",
      "replay_href": "replays/pixels-best-cloudflare--clef-omni.mp4",
      "replay_sha256": "c22f1a04ae9f03b0d111248a6ffedafea7deb5cc34bd65b41f20668e101480da",
      "source_hashes": {
        "requests_sha256": "8d9a0027548c100edbca255fe514d8130de59d132258b9fe0e9a8df1d592ec5a",
        "responses_sha256": "c1050a030bb2d8f2a6bd52a360d4b3a21b7724e4172d6cb05195af9bf7180231",
        "private_replay_sha256": "b3615bd92e17113bbdd0b4c269fd5e3ef2ebcba2a29072400a1f44d67729a722",
        "audit_sha256": "266494870b54a4c4cb4dd0bc54f069bbb6df82924568dd98f822e025997b6f02"
      }
    },
    {
      "id": "pixels-best-inception--mercury-decide",
      "data_class": "development_observation",
      "model_id": "inception/mercury-decide",
      "display_name": "Inception: Mercury Decide",
      "mode": "image",
      "level": "1-1",
      "returned_model": "inception/mercury-decide-20260930",
      "provider": "Inception",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1797.0,
      "progress_pct": 56.29605895546299,
      "observed_progress_pct": 56.29605895546299,
      "native_frames": 740,
      "decisions": 124,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "edgezoom",
        "cadence": 6,
        "history": 2,
        "contract_sha256": "9ff0d844577e90c4c70eb2cf36eafe343bde9a760d3ce132f3e25182189b8b24",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 124
      },
      "trace_href": "traces/pixels-best-inception--mercury-decide.json",
      "trace_sha256": "412a1ee41665c36aa79c9a6c75049583be2f96b3ba1ba82be795fc6c07feb8f2",
      "profile_href": "profiles/pixels-best-inception--mercury-decide.json",
      "profile_sha256": "16fc4455b6e3d170e8645d6367fc6810cac505652edd6fbd8ffbecabce141665",
      "replay_href": "replays/mercury-pixels-partial.mp4",
      "replay_sha256": "600ddef3f4f11b1c6db825642e9ca97c9ec71de87fec0ac3e4810b1f90d986c8",
      "source_hashes": {
        "requests_sha256": "d5c91fc570f58ad0db9c01b468a7bbbc92749aaff3091a19a2a7d6c1d9b3eb2e",
        "responses_sha256": "0407d3b7001e1ac51c05b00f861b481e5b08251cc600629be11b4357a57d6501",
        "private_replay_sha256": "d85302ebd02c8cbd8c779bcdf582b1e96e1b09131eb47cae23c0d34fee3a7ee1",
        "audit_sha256": "941cd6bf7b28644576bad3e8d1c815de601af2131e49a2db0ea23ecbe3bcdd36"
      }
    },
    {
      "id": "pixels-best-liquid--d1-ultra",
      "data_class": "development_observation",
      "model_id": "liquid/d1-ultra",
      "display_name": "LiquidAI: d1-ultra",
      "mode": "image",
      "level": "1-1",
      "returned_model": "liquid/d1-ultra-20261015",
      "provider": "Liquid",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 898.0,
      "progress_pct": 27.491188721563603,
      "observed_progress_pct": 27.491188721563603,
      "native_frames": 486,
      "decisions": 122,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "factoredfirst",
        "cadence": 4,
        "history": 2,
        "contract_sha256": "e13fc9e596f86b83ea3da6856a0582d7dd78f56eff7a33e6b71ee0d276861591",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 122
      },
      "trace_href": "traces/pixels-best-liquid--d1-ultra.json",
      "trace_sha256": "e1611488ecac60151773776650d3501bc5cec45527d92ad0854d7ce554d52ce3",
      "profile_href": "profiles/pixels-best-liquid--d1-ultra.json",
      "profile_sha256": "e92ba40b34d2c579b5cc2df8997fe6f8331501b05b7f1deb6d949ef760cbc1a9",
      "replay_href": "replays/pixels-best-liquid--d1-ultra.mp4",
      "replay_sha256": "fc64b00564a40d10b83737cadbe1c8a3226475665ce011aa42382a7811607a67",
      "source_hashes": {
        "requests_sha256": "a07a29d224bf8cdb40bf88f9e771cbef6caf5c6c51d13b69436a2fd80527723a",
        "responses_sha256": "76fa5a34a8393f0038a90303ab0e6e4013c95366a29f2bb80416346b7356d71b",
        "private_replay_sha256": "8adb22f024e91b71d583818be5e0cfaf83a6ece490b735d02519fdcdab7d9dad",
        "audit_sha256": "ce4b61547bc09f347d2adf2e69d148b658098b85cbe346ae22f2aeabaee5c433"
      }
    },
    {
      "id": "pixels-best-liquid--d1-1",
      "data_class": "development_observation",
      "model_id": "liquid/d1.1",
      "display_name": "LiquidAI: d1.1",
      "mode": "image",
      "level": "1-1",
      "returned_model": "liquid/d1.1-20261027",
      "provider": "Liquid",
      "status": "development_stagnation_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 435.0,
      "progress_pct": 12.656199935917975,
      "observed_progress_pct": 12.656199935917975,
      "native_frames": 720,
      "decisions": 180,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "edgezoom",
        "cadence": 4,
        "history": 2,
        "contract_sha256": "95156eb0af3805bf46603581abdb7c522d9e3d5e0d6124efe19f65264b34d0ff",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 180
      },
      "trace_href": "traces/pixels-best-liquid--d1-1.json",
      "trace_sha256": "82e4a33ee2d5bd9f6efe9a8b14c283791895f84246a6b375d783a128a7d8d4df",
      "profile_href": "profiles/pixels-best-liquid--d1-1.json",
      "profile_sha256": "385ebdb7344332c7332034e4e0f49bbab7ed459b7321cbe895376692c85ad440",
      "replay_href": "replays/pixels-best-liquid--d1-1.mp4",
      "replay_sha256": "205d23f4cf6af49b143e43f0a12a4a5e88153feebdb28db1af2ba902015261b2",
      "source_hashes": {
        "requests_sha256": "b11287caa1992a6d767ca42dda86dd7516292d5a03f7caddd3c97d13200d8ca0",
        "responses_sha256": "078ca5f719723b3831715c826bc822329586c5f7ec420e3300228e3c5e2435ab",
        "private_replay_sha256": "1af0c23b8c699af26136150d403ca61a78be15a88589eca88f6e7d843ef509fe",
        "audit_sha256": "679dd9cfbaf9c7eb6866dd8db14caafab993243a9c0ca5b444694fc5e7d4569e"
      }
    },
    {
      "id": "pixels-best-nace-ai--drex-max",
      "data_class": "development_observation",
      "model_id": "nace-ai/drex-max",
      "display_name": "Nace.AI: Drex Max",
      "mode": "image",
      "level": "1-1",
      "returned_model": "nace-ai/drex-max-20261016",
      "provider": "DeepInfra",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1797.0,
      "progress_pct": 56.29605895546299,
      "observed_progress_pct": 56.29605895546299,
      "native_frames": 1296,
      "decisions": 260,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "edgezoom",
        "cadence": 5,
        "history": 2,
        "contract_sha256": "54eddf5f6c88ba877aca5a9f74c97c0caee5d2c57cf833bf4c163df7352a83ff",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 260
      },
      "trace_href": "traces/pixels-best-nace-ai--drex-max.json",
      "trace_sha256": "7c7ef309235f4808c72bf4973abf9ac94110f86bed57998b92cc4e78b39d9505",
      "profile_href": "profiles/pixels-best-nace-ai--drex-max.json",
      "profile_sha256": "91d4976db65a00d6bb879759d94123c30a6cd39068c87c4a08b1dcefabea6be0",
      "replay_href": "replays/pixels-best-nace-ai--drex-max.mp4",
      "replay_sha256": "c820bd50f25e8c67ce853bcf3caf8615b3d31b6cb6c0cfea4965a0347eba6dd9",
      "source_hashes": {
        "requests_sha256": "eabf714b89be008b427b6f22c28c75621f6689040c75ae3c98217e945bd6e80e",
        "responses_sha256": "b442f541758b79cd8beb6e2e2f538aafd3ec492f8509d991b89b0923517a5c07",
        "private_replay_sha256": "2a028f791c6b11a05374c331fb94c1c3203d557ebd722eca6a9ba1a1ef332569",
        "audit_sha256": "d37f6856d222472732ad8ae0e55c91b0e3bba0fb76447fa6af400f119c4ee513"
      }
    },
    {
      "id": "pixels-best-respan--span-02",
      "data_class": "development_observation",
      "model_id": "respan/span-02",
      "display_name": "Respan: Span-02",
      "mode": "image",
      "level": "1-1",
      "returned_model": "respan/span-02-20261015",
      "provider": "Respan",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 676.0,
      "progress_pct": 20.37808394745274,
      "observed_progress_pct": 20.37808394745274,
      "native_frames": 242,
      "decisions": 18,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "macros",
        "cadence": 8,
        "history": 1,
        "contract_sha256": "233803053de136b284dd443a98a243c1a50dbd1cfda5f466b8af934e71b7aeeb",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": true,
        "requests_audited": 18
      },
      "trace_href": "traces/pixels-best-respan--span-02.json",
      "trace_sha256": "d3f95a2b605f25a0b8c01754bef80d6c57b40dff2a14745006ec1755ab8e54b2",
      "profile_href": "profiles/pixels-best-respan--span-02.json",
      "profile_sha256": "384fc874066806671b2a115b91e0204c55449c87c007bdc19696a0dbe09c7af3",
      "replay_href": "replays/pixels-best-respan--span-02.mp4",
      "replay_sha256": "444b0457a58d8be1cae1b325c1b93f1eb176d4ee60cf669b6744b91da5f884ae",
      "source_hashes": {
        "requests_sha256": "c6b546421a38614b7d81561b22f740fdc0049c11eca995b87d28d53441055388",
        "responses_sha256": "3346ba52bec520efd180dcebe9311505797ad0d488bc71eb59e628d8c30968e6",
        "private_replay_sha256": "7b07fb02b26dce1bcfa5576e1e3fc36de8c77dcf7e053ea806251728ec2ac140",
        "audit_sha256": "25038e6957fbf4022fb0723363796119fbe7839b4d12847086d32d9475e74e55"
      }
    },
    {
      "id": "pixels-best-openai--gpt-6-luna-decisions",
      "data_class": "development_observation",
      "model_id": "openai/gpt-6-luna-decisions",
      "display_name": "OpenAI: GPT-6 Luna Decisions",
      "mode": "image",
      "level": "1-1",
      "returned_model": "openai/gpt-6-luna-decisions-20261006",
      "provider": "OpenAI",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 2470.0,
      "progress_pct": 77.85966036526754,
      "observed_progress_pct": 77.85966036526754,
      "native_frames": 1516,
      "decisions": 304,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "canonical_edges_crop",
        "cadence": 5,
        "history": 1,
        "contract_sha256": "6e48b48a27af25347ef8536800991f15a2343295d7ab13da56eeb19105feb173",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": false,
        "requests_audited": 304,
        "payload_image_byte_audit": "not included in quick native audit; pure pixels builder boundary recorded"
      },
      "trace_href": "traces/pixels-best-openai--gpt-6-luna-decisions.json",
      "trace_sha256": "1f1dc9b01eabc06739c21c8ead665f4bc2f32e0a376399ea1e1c02b074e2ff35",
      "profile_href": "profiles/pixels-best-openai--gpt-6-luna-decisions.json",
      "profile_sha256": "7d638cde291f98ca7441181d6bd97f242b637adae3a7a12190593f448ab92b56",
      "replay_href": "replays/pixels-best-openai--gpt-6-luna-decisions.mp4",
      "replay_sha256": "81b2b88da97c7a3a4781bf0d75fe91cc802cb667dc3609444eda8b65af139403",
      "source_hashes": {
        "requests_sha256": "5a8d58ca37210bff43b938bac01163df5900b34b765e6b0fe13cda8a81e49265",
        "responses_sha256": "41897fbb9deabee1e1b23561d9821f79c7fd15cbedf186e77683aa4ba2273016",
        "private_replay_sha256": "d79cc5f67a153273e9cc1e28c50b19695da74b22c8f1fbfcb8ccb6729cc9a2ff",
        "audit_sha256": "baca3c1665db57a51040e834aa573b1e84046a20e0df83cc3893f1b8eeb4c806"
      }
    },
    {
      "id": "pixels-best-perplexity--pplx-decider-flash-v1-2",
      "data_class": "development_observation",
      "model_id": "perplexity/pplx-decider-flash-v1.2",
      "display_name": "Perplexity: Decider Flash V1.2",
      "mode": "image",
      "level": "1-1",
      "returned_model": "perplexity/pplx-decider-flash-v1.2-20261016",
      "provider": "Perplexity",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1423.0,
      "progress_pct": 44.31272028196091,
      "observed_progress_pct": 44.31272028196091,
      "native_frames": 576,
      "decisions": 144,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "canonical",
        "cadence": 4,
        "history": 2,
        "contract_sha256": "12ca47bfffc8f80df1e1244197ba7d06a057c90a2e427ead2373e401a8cb9532",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": false,
        "requests_audited": 144,
        "payload_image_byte_audit": "not included in quick native audit; pure pixels builder boundary recorded"
      },
      "trace_href": "traces/pixels-best-perplexity--pplx-decider-flash-v1-2.json",
      "trace_sha256": "17eea4204b318481a9f4285cfcb5764ae627a3ac206a116bc6a0ad288c0fd203",
      "profile_href": "profiles/pixels-best-perplexity--pplx-decider-flash-v1-2.json",
      "profile_sha256": "e6728e805cc52cc088acb61a81c9ca8b11dbee1b2ba8aeb6b35bdffbb53d66ca",
      "replay_href": "replays/pixels-best-perplexity--pplx-decider-flash-v1-2.mp4",
      "replay_sha256": "ab227d03d91d867102080bec51ed6aeaeb9699b00ce09c6560f01fe41b977ba4",
      "source_hashes": {
        "requests_sha256": "8565073c49eda438eeb6cff0e7b4d1c7e0ffd264230ba5e568b0f5c8d3363018",
        "responses_sha256": "1d777ead42bb6e93102042b702d9f89240552d26af12c06760c0857a43f00a18",
        "private_replay_sha256": "065f0c6fb8406167a1960422dbc22804c35ff06d340aafbcaa19ebc12b749965",
        "audit_sha256": "dafda53624876dff5d8bf760a948780d031aec1f5902b4b038d49dbd7ddbd937"
      }
    },
    {
      "id": "pixels-best-typesafe--jev-1-13",
      "data_class": "development_observation",
      "model_id": "typesafe/jev-1.13",
      "display_name": "TypeSafe: Jev 1.13",
      "mode": "image",
      "level": "1-1",
      "returned_model": "typesafe/jev-1.13-20260917",
      "provider": "TypeSafe",
      "status": "death",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1126.0,
      "progress_pct": 34.79653957065043,
      "observed_progress_pct": 34.79653957065043,
      "native_frames": 858,
      "decisions": 215,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "canonical",
        "cadence": 4,
        "history": 2,
        "contract_sha256": "12ca47bfffc8f80df1e1244197ba7d06a057c90a2e427ead2373e401a8cb9532",
        "matched_cohort": false,
        "selected_attempt": true,
        "maximum_native_frames": null,
        "development_stall_abort_after_frames": null
      },
      "input_audit": {
        "native_replay_verified": true,
        "privileged_state_sent": false,
        "actual_native_pixels_verified": false,
        "requests_audited": 215,
        "payload_image_byte_audit": "not included in quick native audit; pure pixels builder boundary recorded"
      },
      "trace_href": "traces/pixels-best-typesafe--jev-1-13.json",
      "trace_sha256": "300f8ba1f34eca93e6f19c6a989a8a979a68ec5dc54b31e44dc0191bd6ff46d6",
      "profile_href": "profiles/pixels-best-typesafe--jev-1-13.json",
      "profile_sha256": "2eec936330f5ecda343a8df0ea1064cda29d3708c9b5e6ecc2330424eb65b004",
      "replay_href": "replays/pixels-best-typesafe--jev-1-13.mp4",
      "replay_sha256": "d70244279a61a9e1ea30d54e37e463ba9eec2d1279b930b5d99a6102b62be234",
      "source_hashes": {
        "requests_sha256": "2ea6fe27373c3836c629f8ed28a9f87bd48e20f011d9031b79dab260af92fe3c",
        "responses_sha256": "414e08ff9a07294e0bb4792f4f77e538cda0606ea215c4fd130a1076967d50e2",
        "private_replay_sha256": "08fb25c9a71a751342712a1fede32cb4456c049cd1a91d591e833d24f518a279",
        "audit_sha256": "927d26e7e0f5020b0ee0d9821c735cca48bb6d5ff37e520415777ebd8584333c"
      }
    },
    {
      "id": "pplx-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "perplexity/pplx-decider-v1.2",
      "display_name": "Perplexity: Decider V1.2",
      "returned_model": "perplexity/pplx-decider-v1.2-20261016",
      "provider": "Perplexity",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 2469.0,
      "progress_pct": 77.82761935277155,
      "native_frames": 1302,
      "decisions": 186,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "short edge semantics; explicit A and B choices",
        "adapter": "explicit_ab_joint12",
        "cadence": 7,
        "previous_screens": 1,
        "observation": "current full screen plus 1 past full screen; no RAM",
        "contract_sha256": "c25f2f228f875ed725c1f9044a70d18f08edad1612690b1b555b10f8c7f420fa",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "pplx-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 186,
        "description": "actual current/past screenshots and fixed pixel crops; declared rules/horizons and previous model controls ONLY",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/pplx-pixels-partial.json",
      "trace_sha256": "a3fc585f21f9f774d3ae0bb9b80af48832358158e873d237ab3e72d6b306d79d",
      "replay_href": "replays/pplx-pixels-partial.mp4",
      "replay_sha256": "fe44a6ba6f7da0325cd0d22ba2321cb85b15176a61d8bee46cbd75ca93472f14",
      "source_hashes": {
        "manifest_sha256": "22fca3a0aaf79205431590a3c40c62d92dfbdfecb5aa2b89caa974f824a16396",
        "requests_sha256": "421a36ebbc951e5dc355f200061140ee0291efe2a7ff4faec80191cb42cd04de",
        "responses_sha256": "77ec3925b542085646f6e4d75f391da5d71af0849176591358841b2708e4f739",
        "private_replay_sha256": "124ae679407fb24e38a5fbcca953ad4047f3ca739a51dfc4ae27c6fe6724cdb6",
        "input_audit_sha256": "483d1a8277d5af5ec13c1b50804f8ec0cb25ee09e7d94f85ede915453a4d270c"
      },
      "controls": {
        "adapter": "explicit_ab_joint12",
        "questions": [
          "action"
        ],
        "trace_href": "traces/pplx-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/pplx-pixels-partial.json",
      "profile_sha256": "231d246ba671e176586b212ac4cbc35178c83c1973b497bcbf8fb26621f45fb7"
    },
    {
      "id": "jev-pixels-partial",
      "data_class": "selected_development_examples",
      "model_id": "typesafe/jev-1.14",
      "display_name": "TypeSafe: Jev 1.14",
      "returned_model": "typesafe/jev-1.14-20261014",
      "provider": "TypeSafe",
      "mode": "image",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1661.0,
      "progress_pct": 51.93848125600769,
      "native_frames": 1134,
      "decisions": 189,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "experiment_config": {
        "profile": "long-jump mechanics; four independent button questions",
        "adapter": "independent_buttons",
        "cadence": 6,
        "previous_screens": 1,
        "observation": "2 chronological full screens; no RAM",
        "contract_sha256": "27056f6bead483924a7ba096dfb72dd57148a4d389e3fbdddb45411eee74395b",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "profile_id": "jev-pixels-partial"
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 189,
        "description": "screenshots, declared rules/horizons and previous model controls only",
        "every_request_images_checked": true,
        "native_replay_verified": true
      },
      "trace_href": "traces/jev-pixels-partial.json",
      "trace_sha256": "ab93c6e564f69416614ff71763b232280c7c8b4bc51341ec2e7416d014749909",
      "replay_href": "replays/jev-pixels-partial.mp4",
      "replay_sha256": "e5a690942ff1ffc78a936aada4dfe731ed4156a644913b653580cf9f57d7b828",
      "source_hashes": {
        "manifest_sha256": "f227b23f42c9900b547b829d407aff4319b68ae8ba68483ad852e5209920492d",
        "requests_sha256": "3aae5ed5534d33b2efd52b5b333e74a9f614dafb23b206f1487a7de90930bf1a",
        "responses_sha256": "e018715610a9f14c5e27529177fd7db9a0109f90ec6f3e9f77de2882973c9125",
        "private_replay_sha256": "7a7a14f2559bee48006f2a5328f214344b1d67aee1ce6a9e8a608e8926731625",
        "input_audit_sha256": "090b1d762940a5311dcc24c39aba45a118c9343ef02687944e00d9b8252d57f8"
      },
      "controls": {
        "adapter": "independent_buttons",
        "questions": [
          "horizontal",
          "jump",
          "run",
          "vertical"
        ],
        "trace_href": "traces/jev-pixels-partial.json"
      },
      "selected_attempt_counts": {
        "selected_examples": 1,
        "completions": 0,
        "note": "Selected showcase only, not the number of research attempts or a model clear rate"
      },
      "profile_href": "profiles/jev-pixels-partial.json",
      "profile_sha256": "91400711a83623312598b9b60174fe068ba83cf43fd3dd0a87b95022ed936d18"
    },
    {
      "id": "mammoth-pixels-first-clear",
      "data_class": "selected_development_examples",
      "model_id": "baseten/mammoth-1",
      "display_name": "Baseten: Mammoth 1",
      "returned_model": "baseten/mammoth-1-20261021",
      "provider": "BaseTen",
      "mode": "image",
      "level": "1-1",
      "status": "completed",
      "completed": true,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 3157.0,
      "progress_pct": 100.0,
      "observed_progress_pct": 100.0,
      "native_frames": 1916,
      "decisions": 467,
      "speedrun_score_native_frames": 1916,
      "nominal_game_seconds_to_completion": 31.933333333333334,
      "experiment_config": {
        "profile": "memoryfinish240-v1",
        "adapter": "explicit_ab_joint12",
        "cadence": null,
        "initial_cadence_frames": 6,
        "cadence_mode": "controller-history-triggered native model-selected durations",
        "allowed_selected_durations": [
          2,
          4,
          6,
          8
        ],
        "previous_screens": 1,
        "observation": "current full screen, fixed left/middle crops and past full screen; declared prior model controls only",
        "contract_sha256": "429259115f9aaa9316e87287d8a8a7688bdaf3bf34898e232d296442e251530c",
        "rom_sha256": "ec299b990e8bfee8ba46e3f61d63b2e1ae5b8a2e431de84e2e4bbd692dc53586",
        "seed": 0,
        "life_budget": 1,
        "selected_attempt": true,
        "matched_cohort": false,
        "research_frame_budget": 18000,
        "research_decision_budget": 1600
      },
      "input_audit": {
        "privileged_state_sent": false,
        "requests_audited": 467,
        "actual_pixel_images_audited": 1867,
        "native_replay_verified": true,
        "all_frames_verified": 1916,
        "duration_calls_audited": 255,
        "finish_rule_calls_audited": 7,
        "own_control_history_only": true,
        "independent_audit_pass": true
      },
      "trace_href": "traces/mammoth-pixels-first-clear.json",
      "trace_sha256": "58dadbda3e883f7892aefa5d723e4e8d50adfb2b77d7c4696fba4fcdcb9f3081",
      "profile_href": "profiles/mammoth-pixels-first-clear.json",
      "profile_sha256": "5c8ad5bf2566b8c828d574874b6c69e59093e0ad99841975fa3d5db7d4a7d209",
      "replay_href": "replays/mammoth-pixels-first-clear.mp4",
      "replay_sha256": "b9b1532ed8f97a75a544901849796d8b12db0de5e86f5aa0845cfb56324496b8",
      "replay_includes_initial_still": false,
      "proof_href": "proofs/mammoth-pixels-first-clear.json",
      "source_hashes": {
        "manifest_sha256": "fb31e92a607d29bf7a393065e9cb1fe330898f237b3e513cb93f96e1f2e377fd",
        "requests_sha256": "e3fb3994793fc6fa97858a292ffbed0b6eb3e633d2e2fd916c2b0bf0614f6642",
        "responses_sha256": "b403e00c43e5a9b1f4d7aa15c312192ead620e0c0590a6f1b3dbf94007c1e887",
        "private_replay_sha256": "39627b92d18c4586a4db95626aa863188cbde2c946ec084bdbebb0f57bc508e0",
        "profile_proof_sha256": "7e51d85c01c7b30451c6bbb8a114a2a29f82e76e20935a7910be684d46e85372",
        "input_native_audit_sha256": "a041a012f2dfa749a93f03ed2968c2e34e3b4fac5fb54f1612783f1a51f77123"
      },
      "controls": {
        "adapter": "native simultaneous buttons plus conditional native duration head",
        "trace_href": "traces/mammoth-pixels-first-clear.json"
      }
    }
  ],
  "observation_selection": "Common-profile state attempts, earlier selected pixel partials and newly verified distinct development episodes. Different settings; historical runs retained; no comparable aggregate or model mean.",
  "observation_counts": {
    "state": 22,
    "image": 17
  },
  "technical_diagnostics": [
    {
      "id": "openai-state-json-object-cohort-prefix",
      "data_class": "retrospective_cohort_equivalent_prefix",
      "model_id": "openai/gpt-6-luna-decisions",
      "display_name": "OpenAI: GPT-6 Luna Decisions",
      "returned_model": "openai/gpt-6-luna-decisions-20261006",
      "provider": "OpenAI",
      "mode": "state",
      "level": "1-1",
      "status": "stalled_development_abort",
      "completed": false,
      "death": false,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 596.0,
      "progress_pct": null,
      "observed_progress_pct": 17.81480294777315,
      "final_score": null,
      "native_frames": 880,
      "decisions": 220,
      "api_calls": 220,
      "api_errors": 0,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "comparable_aggregate": null,
      "replaces_original_observation": false,
      "original_observation_id": "state-c4-openai--gpt-6-luna-decisions",
      "trace_href": "traces/openai-state-json-object-cohort-prefix.json",
      "trace_sha256": "36d97cda5205a4a1e938c013d1dd9d2a6d15ee57e8a66ae71e98665c2447a924",
      "profile_href": "profiles/openai-state-json-object-cohort-prefix.json",
      "profile_sha256": "4d553480467d088c21a4a6d70319c21a4c000896e89228448633ee20e1dff692",
      "replay_href": "replays/openai-state-json-object-cohort-prefix.mp4",
      "source_hashes": {
        "source_episode_sha256": "b398e376aef9949c2faf35d2930804c90b00f130311f630643d9662b57a174c7",
        "frozen_manifest_sha256": "8f9b42264d0a8524b9240f954994f1427192ebbe21fcf7e0a2617342da2a220a",
        "technical_report_sha256": "67b2f54f67763d3fd9a1678d9bb7f5d8470f9fff13c6e31085d268071099cf29"
      },
      "experiment_config": {
        "profile": "joint-visible-v1",
        "cadence": 4,
        "state_serialization": "json_object",
        "questions_rules_and_evidence_text_unchanged": true,
        "matched_cohort": false,
        "retrospective": true,
        "contract_sha256": "bb26221092a91e20c2eee867122e9d821234aba7879ef0244031c76dc500810d"
      },
      "replay_sha256": "6ed1848b8bbfcd6c80ecd1acf7ea9b255cbdcb26e0eb8be000cccdbbec78ea2f"
    },
    {
      "id": "openai-state-json-object-full-diagnostic",
      "data_class": "technical_serialization_diagnostic",
      "model_id": "openai/gpt-6-luna-decisions",
      "display_name": "OpenAI: GPT-6 Luna Decisions",
      "returned_model": "openai/gpt-6-luna-decisions-20261006",
      "provider": "OpenAI",
      "mode": "state",
      "level": "1-1",
      "status": "game_over",
      "completed": false,
      "death": true,
      "verified": true,
      "start_x": 40,
      "goal_x": 3161,
      "furthest_valid_x": 1125.0,
      "progress_pct": 34.764498558154436,
      "observed_progress_pct": 34.764498558154436,
      "final_score": null,
      "native_frames": 2059,
      "decisions": 515,
      "api_calls": 515,
      "api_errors": 0,
      "speedrun_score_native_frames": null,
      "nominal_game_seconds_to_completion": null,
      "comparable_aggregate": null,
      "replaces_original_observation": false,
      "original_observation_id": "state-c4-openai--gpt-6-luna-decisions",
      "trace_href": "traces/openai-state-json-object-full-diagnostic.json",
      "trace_sha256": "e289ceb2d5b7f8ed9501a5bc109d646fa7206e5ea0e8f060ffa41f5a2c5f99bc",
      "profile_href": "profiles/openai-state-json-object-full-diagnostic.json",
      "profile_sha256": "6970eb6e9110ac69502a236c831de8aac16c95688c6c45d4e0d8487461f889a6",
      "replay_href": "replays/openai-state-json-object-full-diagnostic.mp4",
      "source_hashes": {
        "source_episode_sha256": "b398e376aef9949c2faf35d2930804c90b00f130311f630643d9662b57a174c7",
        "frozen_manifest_sha256": "8f9b42264d0a8524b9240f954994f1427192ebbe21fcf7e0a2617342da2a220a",
        "technical_report_sha256": "67b2f54f67763d3fd9a1678d9bb7f5d8470f9fff13c6e31085d268071099cf29"
      },
      "experiment_config": {
        "profile": "joint-visible-v1",
        "cadence": 4,
        "state_serialization": "json_object",
        "questions_rules_and_evidence_text_unchanged": true,
        "matched_cohort": false,
        "retrospective": true,
        "contract_sha256": "bb26221092a91e20c2eee867122e9d821234aba7879ef0244031c76dc500810d"
      },
      "replay_sha256": "0a050e54724b0a912605464b9adf04d6c5a2e87e7b0efc90b751346f7862a85d"
    }
  ],
  "technical_diagnostic_report_href": "diagnostics/openai-state-serialization.json"
}
