{
 "chart": "make_eval_usage/make_eval_usage",
 "title": "Resource use in Making METR tasks",
 "subtitle": "The task-making agent is given 3 test environments, and token budgets for itself, a frontier model, and a lagging model",
 "panels": [
  {
   "title": "",
   "x": {"label": "Elapsed time (hours)", "unit": "", "scale": "linear"},
   "y": {"label": "", "unit": "", "scale": "linear"},
   "tables": [
    {
     "name": "phases and test runs",
     "rows": [
      {"model": "GPT-4o", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.113},
      {"model": "GPT-4o", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.113, "end_hours": 0.134},
      {"model": "GPT-4o", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 0.113, "end_hours": 0.132},
      {"model": "GPT-4o", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.118},
      {"model": "GPT-4o", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.118, "end_hours": 0.137},
      {"model": "GPT-4o", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 0.118, "end_hours": 0.135},
      {"model": "GPT-4o", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.045},
      {"model": "GPT-4o", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.045, "end_hours": 0.069},
      {"model": "GPT-4o", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 0.045, "end_hours": 0.068},
      {"model": "o1", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.03},
      {"model": "o1", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.03, "end_hours": 0.032},
      {"model": "o1", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 0.03, "end_hours": 0.031},
      {"model": "o1", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.075},
      {"model": "o1", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.075, "end_hours": 0.077},
      {"model": "o1", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 0.075, "end_hours": 0.075},
      {"model": "o1", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.019},
      {"model": "o1", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.019, "end_hours": 0.022},
      {"model": "o1", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 0.019, "end_hours": 0.021},
      {"model": "GPT-5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.468},
      {"model": "GPT-5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.468, "end_hours": 0.473},
      {"model": "GPT-5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 0.468, "end_hours": 0.469},
      {"model": "GPT-5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.361},
      {"model": "GPT-5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.361, "end_hours": 0.364},
      {"model": "GPT-5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 0.361, "end_hours": 0.362},
      {"model": "GPT-5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 0.323},
      {"model": "GPT-5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 0.323, "end_hours": 0.327},
      {"model": "GPT-5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 0.323, "end_hours": 0.324},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 3.909},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 3.909, "end_hours": 4.244},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.192, "end_hours": 0.315},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.383, "end_hours": 0.426},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 0.437, "end_hours": 0.49},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.652, "end_hours": 0.745},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.838, "end_hours": 0.906},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 0.977, "end_hours": 1.161},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.245, "end_hours": 1.3},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.408, "end_hours": 1.51},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.528, "end_hours": 1.593},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.98, "end_hours": 2.012},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.124, "end_hours": 2.181},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 2.242, "end_hours": 2.313},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.362, "end_hours": 2.461},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.495, "end_hours": 2.594},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 2.651, "end_hours": 2.911},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.931, "end_hours": 3.212},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.938, "end_hours": 2.98},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.258, "end_hours": 3.509},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 3.532, "end_hours": 3.852},
      {"model": "Claude Opus 4.5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 3.909, "end_hours": 4.231},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 8.114},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 8.114, "end_hours": 8.201},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.18, "end_hours": 0.233},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.295, "end_hours": 0.379},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.387, "end_hours": 0.508},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.556, "end_hours": 0.662},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.669, "end_hours": 0.915},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.952, "end_hours": 1.053},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.062, "end_hours": 1.323},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.367, "end_hours": 1.764},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 1.804, "end_hours": 1.919},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.928, "end_hours": 2.298},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.422, "end_hours": 2.527},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.635, "end_hours": 2.852},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.934, "end_hours": 3.038},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.046, "end_hours": 3.258},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.37, "end_hours": 3.515},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.371, "end_hours": 3.543},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 3.42, "end_hours": 3.76},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.989, "end_hours": 4.331},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.99, "end_hours": 4.564},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 3.997, "end_hours": 4.32},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.38, "end_hours": 4.764},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 4.38, "end_hours": 4.654},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.855, "end_hours": 5.2},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.855, "end_hours": 5.1},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 4.856, "end_hours": 5.207},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.232, "end_hours": 5.704},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.233, "end_hours": 5.443},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.729, "end_hours": 6.144},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.729, "end_hours": 6.153},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.199, "end_hours": 6.344},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 6.199, "end_hours": 8.114},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.864, "end_hours": 6.906},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 6.864, "end_hours": 8.114},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.989, "end_hours": 7.147},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.209, "end_hours": 7.299},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.31, "end_hours": 7.357},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.378, "end_hours": 7.414},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 7.433, "end_hours": 8.114},
      {"model": "Claude Opus 4.5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 8.114, "end_hours": 8.167},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 2.089},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 2.089, "end_hours": 2.177},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.228, "end_hours": 0.259},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.362, "end_hours": 0.439},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.493, "end_hours": 0.572},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.604, "end_hours": 0.69},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.713, "end_hours": 0.958},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.985, "end_hours": 1.096},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.159, "end_hours": 1.252},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.313, "end_hours": 1.396},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 1.418, "end_hours": 1.498},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.616, "end_hours": 1.722},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.736, "end_hours": 1.81},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 1.83, "end_hours": 1.892},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.949, "end_hours": 2.025},
      {"model": "Claude Opus 4.5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 2.089, "end_hours": 2.168},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 3.217},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 3.217, "end_hours": 3.322},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.481, "end_hours": 0.531},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.721, "end_hours": 0.772},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.991, "end_hours": 1.047},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.334, "end_hours": 1.464},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.555, "end_hours": 1.655},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.818, "end_hours": 2.228},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.301, "end_hours": 2.395},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 2.405, "end_hours": 2.473},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.507, "end_hours": 3.009},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 3.045, "end_hours": 3.167},
      {"model": "Claude Opus 4.6", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 3.217, "end_hours": 3.312},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 5.341},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 5.341, "end_hours": 6.365},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.454, "end_hours": 0.543},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.83, "end_hours": 0.97},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.836, "end_hours": 1.014},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.202, "end_hours": 1.47},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.991, "end_hours": 2.058},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.386, "end_hours": 2.555},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.6, "end_hours": 2.804},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.922, "end_hours": 2.972},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.104, "end_hours": 3.62},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.104, "end_hours": 3.184},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.755, "end_hours": 4.375},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 4.495, "end_hours": 4.604},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.627, "end_hours": 5.274},
      {"model": "Claude Opus 4.6", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 5.341, "end_hours": 6.354},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 4.166},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 4.166, "end_hours": 4.61},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.591, "end_hours": 0.626},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.832, "end_hours": 0.88},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 0.939, "end_hours": 1.11},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.354, "end_hours": 1.548},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.729, "end_hours": 1.772},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.891, "end_hours": 1.941},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.019, "end_hours": 2.102},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 2.194, "end_hours": 2.33},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.436, "end_hours": 2.79},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.866, "end_hours": 3.248},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.302, "end_hours": 3.669},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: lagging", "test_environment": 3, "start_hours": 3.303, "end_hours": 3.339},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 3.725, "end_hours": 4.096},
      {"model": "Claude Opus 4.6", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 4.166, "end_hours": 4.595},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 1.707},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 1.707, "end_hours": 1.899},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.277, "end_hours": 0.37},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.281, "end_hours": 0.396},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.634, "end_hours": 0.821},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.637, "end_hours": 0.827},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.871, "end_hours": 0.978},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.875, "end_hours": 1.211},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.291, "end_hours": 1.502},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.295, "end_hours": 1.442},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.531, "end_hours": 1.685},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.534, "end_hours": 1.688},
      {"model": "GPT-5.4", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 1.707, "end_hours": 1.876},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 3.57},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 3.57, "end_hours": 4.691},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.17, "end_hours": 0.205},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.173, "end_hours": 0.182},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.883, "end_hours": 1.199},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.886, "end_hours": 1.289},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.404, "end_hours": 1.591},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.408, "end_hours": 1.874},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.939, "end_hours": 2.074},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.943, "end_hours": 2.097},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.209, "end_hours": 2.384},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.213, "end_hours": 2.561},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.586, "end_hours": 2.748},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.59, "end_hours": 2.702},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.794, "end_hours": 3.131},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.798, "end_hours": 3.524},
      {"model": "GPT-5.4", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 3.57, "end_hours": 4.675},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 1.771},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 1.771, "end_hours": 1.913},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.232, "end_hours": 0.339},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.24, "end_hours": 0.299},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.37, "end_hours": 0.468},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.582, "end_hours": 0.709},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.759, "end_hours": 0.933},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.121, "end_hours": 1.575},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.629, "end_hours": 1.717},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.632, "end_hours": 1.679},
      {"model": "GPT-5.4", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 1.771, "end_hours": 1.898},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 6.1},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 6.1, "end_hours": 6.249},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.388, "end_hours": 0.46},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.676, "end_hours": 0.787},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.887, "end_hours": 1.038},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.136, "end_hours": 1.26},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 1, "start_hours": 1.288, "end_hours": 1.434},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.57, "end_hours": 1.725},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.782, "end_hours": 1.919},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.972, "end_hours": 2.091},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.188, "end_hours": 2.287},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.355, "end_hours": 2.439},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.595, "end_hours": 2.708},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.602, "end_hours": 2.712},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.821, "end_hours": 2.945},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.968, "end_hours": 3.075},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.123, "end_hours": 3.217},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.255, "end_hours": 3.342},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.386, "end_hours": 3.475},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 1, "start_hours": 3.487, "end_hours": 3.607},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.672, "end_hours": 3.824},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.905, "end_hours": 4.059},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.117, "end_hours": 4.274},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.521, "end_hours": 4.637},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.88, "end_hours": 4.992},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.032, "end_hours": 5.144},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 1, "start_hours": 5.162, "end_hours": 5.29},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.356, "end_hours": 5.441},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.361, "end_hours": 5.419},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.541, "end_hours": 5.68},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.782, "end_hours": 5.928},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.783, "end_hours": 5.943},
      {"model": "Claude Opus 4.7", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 6.1, "end_hours": 6.226},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 12.013},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 12.013, "end_hours": 12.603},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.324, "end_hours": 0.403},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.471, "end_hours": 0.542},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.641, "end_hours": 0.751},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 0.798, "end_hours": 0.899},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.403, "end_hours": 1.524},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 1.564, "end_hours": 1.733},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.848, "end_hours": 1.952},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.083, "end_hours": 2.303},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.396, "end_hours": 2.522},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.869, "end_hours": 3.25},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 3.268, "end_hours": 3.401},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.899, "end_hours": 4.372},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.589, "end_hours": 5.006},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.021, "end_hours": 5.29},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 5.3, "end_hours": 5.833},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 5.844, "end_hours": 6.049},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 6.475, "end_hours": 7.386},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.779, "end_hours": 8.444},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 8.829, "end_hours": 9.017},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.366, "end_hours": 9.928},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.242, "end_hours": 11.971},
      {"model": "Claude Opus 4.7", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 12.013, "end_hours": 12.59},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 7.412},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 7.412, "end_hours": 7.703},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.456, "end_hours": 0.572},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.817, "end_hours": 1.185},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.38, "end_hours": 1.603},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.39, "end_hours": 1.545},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.718, "end_hours": 1.946},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.045, "end_hours": 2.31},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.053, "end_hours": 2.183},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.372, "end_hours": 2.571},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.623, "end_hours": 2.851},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.625, "end_hours": 3.018},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.13, "end_hours": 3.36},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.132, "end_hours": 3.637},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.826, "end_hours": 4.144},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.207, "end_hours": 4.486},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.209, "end_hours": 4.433},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.734, "end_hours": 5.286},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.737, "end_hours": 5.008},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.763, "end_hours": 6.106},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.764, "end_hours": 6.136},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.264, "end_hours": 6.624},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.837, "end_hours": 7.133},
      {"model": "Claude Opus 4.7", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 7.412, "end_hours": 7.694},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 7.32},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 7.32, "end_hours": 8.393},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.265, "end_hours": 0.746},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.072, "end_hours": 1.775},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.781, "end_hours": 3.006},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.117, "end_hours": 7.264},
      {"model": "GPT-5.5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 7.32, "end_hours": 8.383},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 6.491},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 6.491, "end_hours": 7.855},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.308, "end_hours": 0.647},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.252, "end_hours": 1.804},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.226, "end_hours": 3.108},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.824, "end_hours": 4.46},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.694, "end_hours": 6.433},
      {"model": "GPT-5.5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 6.491, "end_hours": 7.838},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 16.753},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 16.753, "end_hours": 19.191},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.143, "end_hours": 0.307},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.693, "end_hours": 1.803},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.925, "end_hours": 4.968},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.269, "end_hours": 8.271},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 10.062, "end_hours": 11.398},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.109, "end_hours": 14.529},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 15.5, "end_hours": 16.74},
      {"model": "GPT-5.5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 16.753, "end_hours": 19.146},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 16.186},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 16.186, "end_hours": 17.699},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.854, "end_hours": 1.605},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 1.653, "end_hours": 2.827},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 1.654, "end_hours": 2.264},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.959, "end_hours": 2.967},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.114, "end_hours": 3.174},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.19, "end_hours": 3.199},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.251, "end_hours": 5.36},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 3.252, "end_hours": 5.114},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 3.252, "end_hours": 8.529},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.133, "end_hours": 8.373},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 8.558, "end_hours": 10.39},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 8.559, "end_hours": 10.938},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.171, "end_hours": 12.369},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 11.172, "end_hours": 13.235},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.483, "end_hours": 13.52},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.547, "end_hours": 13.93},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 13.553, "end_hours": 14.297},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 14.389, "end_hours": 14.395},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "test run: lagging", "test_environment": 1, "start_hours": 14.705, "end_hours": 15.696},
      {"model": "Claude Opus 4.8", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 16.186, "end_hours": 17.685},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 20.534},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 20.534, "end_hours": 23.063},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.687, "end_hours": 3.061},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.225, "end_hours": 4.714},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.226, "end_hours": 3.988},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.895, "end_hours": 7.453},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.896, "end_hours": 5.694},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.944, "end_hours": 10.052},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 10.225, "end_hours": 11.894},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.151, "end_hours": 12.381},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.523, "end_hours": 14.95},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 15.088, "end_hours": 17.726},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 15.898, "end_hours": 16.987},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 17.76, "end_hours": 20.135},
      {"model": "Claude Opus 4.8", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 20.534, "end_hours": 23.028},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 22.535},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 22.535, "end_hours": 25.594},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.824, "end_hours": 2.162},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.326, "end_hours": 2.959},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.631, "end_hours": 3.972},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.023, "end_hours": 5.011},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.023, "end_hours": 4.786},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 5.209, "end_hours": 6.566},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.842, "end_hours": 9.428},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 6.842, "end_hours": 8.675},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 6.843, "end_hours": 8.557},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.493, "end_hours": 10.954},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.023, "end_hours": 11.337},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 11.024, "end_hours": 12.037},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.136, "end_hours": 14.801},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 12.137, "end_hours": 12.533},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 14.871, "end_hours": 16.117},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 14.871, "end_hours": 16.877},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 16.949, "end_hours": 19.361},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 16.95, "end_hours": 19.396},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 19.547, "end_hours": 21.373},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 19.547, "end_hours": 20.116},
      {"model": "Claude Opus 4.8", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 22.535, "end_hours": 25.526},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 12.059},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 12.059, "end_hours": 13.032},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.76, "end_hours": 1.965},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.102, "end_hours": 4.394},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.11, "end_hours": 4.129},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.377, "end_hours": 6.3},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.39, "end_hours": 5.931},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.607, "end_hours": 7.552},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 6.608, "end_hours": 7.41},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.678, "end_hours": 8.945},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 7.678, "end_hours": 8.469},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 8.977, "end_hours": 9.487},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 8.978, "end_hours": 9.342},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 10.012, "end_hours": 10.565},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 10.012, "end_hours": 10.251},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.004, "end_hours": 12.023},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 11.004, "end_hours": 11.844},
      {"model": "Claude Fable 5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 12.059, "end_hours": 12.968},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 40.3},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 40.3, "end_hours": 42.432},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.573, "end_hours": 1.589},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.706, "end_hours": 1.742},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.768, "end_hours": 1.8},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.855, "end_hours": 3.351},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.855, "end_hours": 2.933},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.392, "end_hours": 4.975},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 3.393, "end_hours": 5.807},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.409, "end_hours": 4.888},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.909, "end_hours": 7.322},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.91, "end_hours": 6.206},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.385, "end_hours": 9.64},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 7.385, "end_hours": 7.827},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 9.76, "end_hours": 9.799},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.816, "end_hours": 11.924},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 9.817, "end_hours": 10.535},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.975, "end_hours": 12.577},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 11.976, "end_hours": 13.013},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 11.976, "end_hours": 12.914},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.321, "end_hours": 13.351},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.371, "end_hours": 13.401},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 13.413, "end_hours": 14.914},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 13.413, "end_hours": 14.097},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 15.236, "end_hours": 15.279},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 15.293, "end_hours": 17.227},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 15.293, "end_hours": 16.653},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 17.277, "end_hours": 17.984},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 17.277, "end_hours": 18.5},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 19.898, "end_hours": 21.47},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 19.899, "end_hours": 20.834},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 20.871, "end_hours": 22.482},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 21.806, "end_hours": 22.552},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 22.767, "end_hours": 24.293},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 22.768, "end_hours": 23.535},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 23.728, "end_hours": 24.002},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 24.706, "end_hours": 25.214},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 24.714, "end_hours": 25.684},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 25.762, "end_hours": 26.818},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 25.762, "end_hours": 26.309},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 26.735, "end_hours": 28.472},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 27.215, "end_hours": 27.516},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 27.701, "end_hours": 29.61},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 28.667, "end_hours": 29.974},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 29.628, "end_hours": 30.014},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 30.19, "end_hours": 30.719},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 30.19, "end_hours": 31.125},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 31.152, "end_hours": 32.297},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 31.152, "end_hours": 32.258},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 32.6, "end_hours": 33.52},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 32.601, "end_hours": 34.016},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 33.571, "end_hours": 34.553},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 35.009, "end_hours": 35.669},
      {"model": "Claude Fable 5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 40.3, "end_hours": 42.396},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 20.882},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 20.882, "end_hours": 21.767},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.554, "end_hours": 0.708},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.128, "end_hours": 1.988},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.712, "end_hours": 1.914},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 3, "start_hours": 1.922, "end_hours": 2.806},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.321, "end_hours": 3.215},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.295, "end_hours": 4.391},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.833, "end_hours": 5.851},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.834, "end_hours": 6.141},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.9, "end_hours": 6.978},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.441, "end_hours": 8.667},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 7.442, "end_hours": 8.526},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.441, "end_hours": 10.435},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.636, "end_hours": 12.609},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 12.854, "end_hours": 13.153},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 14.188, "end_hours": 14.787},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 15.696, "end_hours": 16.0},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 16.012, "end_hours": 17.028},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 17.203, "end_hours": 18.609},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 19.654, "end_hours": 20.572},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 19.655, "end_hours": 20.616},
      {"model": "Claude Fable 5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 20.882, "end_hours": 21.744},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 9.086},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 9.086, "end_hours": 9.772},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.317, "end_hours": 1.025},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.321, "end_hours": 1.166},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.564, "end_hours": 1.249},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.536, "end_hours": 2.211},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.539, "end_hours": 2.593},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.866, "end_hours": 3.577},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.966, "end_hours": 4.717},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 4.989, "end_hours": 5.629},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 5.847, "end_hours": 6.689},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.851, "end_hours": 8.442},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.653, "end_hours": 8.243},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 9.086, "end_hours": 9.743},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 3.156},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 3.156, "end_hours": 3.746},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.341, "end_hours": 1.144},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.345, "end_hours": 0.923},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.545, "end_hours": 0.76},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.37, "end_hours": 2.371},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.375, "end_hours": 2.299},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 2.378, "end_hours": 3.114},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 3.156, "end_hours": 3.733},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 5.019},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 5.019, "end_hours": 9.549},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.339, "end_hours": 0.563},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.343, "end_hours": 0.572},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.759, "end_hours": 0.999},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.763, "end_hours": 1.008},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.568, "end_hours": 2.096},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.572, "end_hours": 2.07},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.248, "end_hours": 3.454},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 2.251, "end_hours": 2.613},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.484, "end_hours": 4.699},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.739, "end_hours": 4.91},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 5.019, "end_hours": 9.501},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 9.131},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 9.131, "end_hours": 9.966},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.292, "end_hours": 1.453},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.494, "end_hours": 2.26},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.382, "end_hours": 2.783},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.159, "end_hours": 3.607},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.627, "end_hours": 4.966},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.59, "end_hours": 6.546},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.669, "end_hours": 7.966},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 8.315, "end_hours": 8.352},
      {"model": "Claude Opus 5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 9.131, "end_hours": 9.966},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 14.017},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 14.017, "end_hours": 14.905},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.083, "end_hours": 0.105},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.907, "end_hours": 1.018},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 1.987, "end_hours": 2.931},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 3, "start_hours": 1.987, "end_hours": 2.962},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.533, "end_hours": 4.517},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.533, "end_hours": 4.211},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.414, "end_hours": 5.819},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 5.414, "end_hours": 6.571},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 5.922, "end_hours": 6.866},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 7.087, "end_hours": 7.504},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 7.088, "end_hours": 8.413},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 7.594, "end_hours": 8.079},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 8.116, "end_hours": 8.948},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 9.066, "end_hours": 10.013},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 10.19, "end_hours": 10.591},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 10.191, "end_hours": 10.499},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 10.853, "end_hours": 11.772},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 10.853, "end_hours": 12.441},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.944, "end_hours": 13.861},
      {"model": "Claude Opus 5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 14.017, "end_hours": 14.887},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 17.88},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 17.88, "end_hours": 19.17},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.742, "end_hours": 0.75},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.727, "end_hours": 2.989},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.728, "end_hours": 2.836},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.766, "end_hours": 1.773},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.823, "end_hours": 1.834},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.409, "end_hours": 8.011},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 6.41, "end_hours": 8.431},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 8.376, "end_hours": 9.524},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.584, "end_hours": 11.05},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 9.584, "end_hours": 10.735},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 9.594, "end_hours": 10.129},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.794, "end_hours": 13.147},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 11.794, "end_hours": 13.11},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 13.748, "end_hours": 15.143},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 15.478, "end_hours": 16.265},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 16.384, "end_hours": 17.469},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 16.385, "end_hours": 17.004},
      {"model": "Claude Opus 5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 17.88, "end_hours": 19.141},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 41.219},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 41.219, "end_hours": 44.715},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.209, "end_hours": 0.218},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.8, "end_hours": 3.258},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.627, "end_hours": 10.99},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 6.628, "end_hours": 9.872},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 6.628, "end_hours": 9.332},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 10.684, "end_hours": 14.546},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 11.53, "end_hours": 15.219},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 15.079, "end_hours": 19.586},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 21.817, "end_hours": 26.057},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 21.817, "end_hours": 25.56},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 25.968, "end_hours": 29.285},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 25.969, "end_hours": 30.029},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 27.513, "end_hours": 30.132},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 31.072, "end_hours": 34.707},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 31.073, "end_hours": 35.128},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 31.073, "end_hours": 35.106},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 35.078, "end_hours": 38.59},
      {"model": "Claude Fable 5.1", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 41.219, "end_hours": 44.625},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 28.263},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 28.263, "end_hours": 30.182},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.535, "end_hours": 1.641},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.748, "end_hours": 1.782},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.691, "end_hours": 7.796},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 6.691, "end_hours": 10.079},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 11.1, "end_hours": 13.401},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 11.1, "end_hours": 13.032},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 17.24, "end_hours": 19.459},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 17.241, "end_hours": 20.899},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 21.65, "end_hours": 23.504},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 21.65, "end_hours": 25.481},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 23.671, "end_hours": 25.37},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 23.672, "end_hours": 25.091},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 25.814, "end_hours": 27.898},
      {"model": "Claude Fable 5.1", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 28.263, "end_hours": 30.116},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 29.507},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 29.507, "end_hours": 33.211},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 6.081, "end_hours": 7.201},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.225, "end_hours": 16.881},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 12.226, "end_hours": 14.435},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 2, "start_hours": 14.77, "end_hours": 19.123},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 19.262, "end_hours": 25.473},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 19.262, "end_hours": 23.937},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 19.27, "end_hours": 21.996},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 24.214, "end_hours": 29.04},
      {"model": "Claude Fable 5.1", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 29.507, "end_hours": 33.17},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 5.216},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 5.216, "end_hours": 5.988},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.29, "end_hours": 1.166},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.392, "end_hours": 0.864},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.622, "end_hours": 1.744},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.409, "end_hours": 2.146},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.494, "end_hours": 2.069},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 1.85, "end_hours": 2.536},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.257, "end_hours": 2.993},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 3.636, "end_hours": 4.562},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 3.636, "end_hours": 4.519},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "test run: lagging", "test_environment": 1, "start_hours": 3.637, "end_hours": 5.007},
      {"model": "GPT-6 Astra", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 5.216, "end_hours": 5.87},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 3.797},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 3.797, "end_hours": 6.042},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.218, "end_hours": 2.135},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.239, "end_hours": 1.191},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.933, "end_hours": 3.209},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 1.305, "end_hours": 1.921},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.024, "end_hours": 3.507},
      {"model": "GPT-6 Astra", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 3.797, "end_hours": 5.987},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 6.965},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 6.965, "end_hours": 9.721},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.189, "end_hours": 3.406},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 0.2, "end_hours": 4.819},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 0.246, "end_hours": 0.488},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 2.759, "end_hours": 6.724},
      {"model": "GPT-6 Astra", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 6.965, "end_hours": 9.658},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 27.28},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 27.28, "end_hours": 30.822},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.137, "end_hours": 0.15},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.991, "end_hours": 6.767},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 4.0, "end_hours": 9.258},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 6.734, "end_hours": 9.163},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.641, "end_hours": 11.26},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 9.642, "end_hours": 11.901},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 9.642, "end_hours": 12.071},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 14.51, "end_hours": 18.483},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 14.511, "end_hours": 17.812},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 14.512, "end_hours": 20.548},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 2, "start_hours": 17.843, "end_hours": 20.06},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 18.777, "end_hours": 22.06},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 2, "start_hours": 20.106, "end_hours": 25.909},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 3, "start_hours": 20.766, "end_hours": 23.166},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: frontier", "test_environment": 1, "start_hours": 23.274, "end_hours": 26.431},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "test run: lagging", "test_environment": 3, "start_hours": 23.274, "end_hours": 27.127},
      {"model": "Claude Opus 5.5", "run": "Run 1", "activity": "final test run", "test_environment": null, "start_hours": 27.28, "end_hours": 30.781},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 16.942},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 16.942, "end_hours": 17.973},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 0.215, "end_hours": 0.225},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 1.676, "end_hours": 1.838},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 2.179, "end_hours": 2.191},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 2.18, "end_hours": 2.193},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.516, "end_hours": 5.184},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.516, "end_hours": 5.832},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 5.691, "end_hours": 5.712},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 5.691, "end_hours": 6.833},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 1, "start_hours": 6.206, "end_hours": 7.888},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 7.643, "end_hours": 7.69},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 7.644, "end_hours": 8.755},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 7.709, "end_hours": 9.329},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 8.8, "end_hours": 10.05},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 9.331, "end_hours": 10.558},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 9.357, "end_hours": 10.645},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 10.76, "end_hours": 10.813},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 10.835, "end_hours": 11.996},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 10.836, "end_hours": 11.499},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 10.847, "end_hours": 11.947},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 12.313, "end_hours": 13.388},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 3, "start_hours": 12.314, "end_hours": 13.448},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: lagging", "test_environment": 2, "start_hours": 12.314, "end_hours": 15.373},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 1, "start_hours": 15.408, "end_hours": 16.613},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "test run: frontier", "test_environment": 2, "start_hours": 16.84, "end_hours": 16.902},
      {"model": "Claude Opus 5.5", "run": "Run 2", "activity": "final test run", "test_environment": null, "start_hours": 16.942, "end_hours": 17.928},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "task development", "test_environment": null, "start_hours": 0.0, "end_hours": 32.681},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "post-submit (final test run)", "test_environment": null, "start_hours": 32.681, "end_hours": 36.22},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 1, "start_hours": 2.135, "end_hours": 2.837},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 3.875, "end_hours": 7.477},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 3.877, "end_hours": 4.896},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 13.391, "end_hours": 15.853},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 13.392, "end_hours": 14.85},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 16.172, "end_hours": 19.375},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 16.173, "end_hours": 20.102},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 18.084, "end_hours": 21.719},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 20.195, "end_hours": 23.108},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 20.196, "end_hours": 23.551},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 24.014, "end_hours": 27.633},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 24.015, "end_hours": 27.704},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 1, "start_hours": 28.25, "end_hours": 32.147},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: frontier", "test_environment": 3, "start_hours": 28.251, "end_hours": 32.491},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "test run: lagging", "test_environment": 2, "start_hours": 28.262, "end_hours": 29.859},
      {"model": "Claude Opus 5.5", "run": "Run 3", "activity": "final test run", "test_environment": null, "start_hours": 32.681, "end_hours": 36.158}
     ]
    }
   ]
  },
  {
   "title": "Task-making model 100M budget",
   "x": {"label": "tokens", "unit": "", "scale": "linear"},
   "y": {"label": "", "unit": "", "scale": "linear"},
   "tables": [
    {
     "name": "token use",
     "rows": [
      {"model": "GPT-4o", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 325900, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 348244, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 84687, "token_budget": 100000000},
      {"model": "o1", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 7579, "token_budget": 100000000},
      {"model": "o1", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 64066, "token_budget": 100000000},
      {"model": "o1", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 8289, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 778367, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 376918, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 334217, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 16569697, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 18815022, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 9141372, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 21499210, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 32834058, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 30188773, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 14844300, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 11258682, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 12023054, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 86271923, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 82993019, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 56056787, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 4702448, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 9248416, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 14028490, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 63391194, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 81364057, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 58370640, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 83130707, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 74495587, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 78283920, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 43622761, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 29650209, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 43189535, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 100281984, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 85808290, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 96385167, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 39601354, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 92078049, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 85070225, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 10666193, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 3953988, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 12156905, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 1", "token_pool": "Task-making model", "tokens_used": 81684225, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 2", "token_pool": "Task-making model", "tokens_used": 82106266, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 3", "token_pool": "Task-making model", "tokens_used": 65577119, "token_budget": 100000000}
     ]
    }
   ]
  },
  {
   "title": "Frontier test model (Opus 5.5) 100M budget",
   "x": {"label": "tokens", "unit": "", "scale": "linear"},
   "y": {"label": "", "unit": "", "scale": "linear"},
   "tables": [
    {
     "name": "token use",
     "rows": [
      {"model": "GPT-4o", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 2943423, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 25364824, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 2835453, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 2341927, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 18322734, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 9279659, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 4961793, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 3941341, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 10457862, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 19056293, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 17118769, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 25694476, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 24438958, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 17868784, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 32157319, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 82684084, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 59945953, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 97536514, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 48187791, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 86093648, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 97710021, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 69500801, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 30396704, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 10488797, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 74844290, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 71022568, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 67679037, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 99484494, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 56204623, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 45841596, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 67177056, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 27679222, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 19986793, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 1", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 97563320, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 2", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 96309565, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 3", "token_pool": "Frontier test model (Opus 5.5)", "tokens_used": 96492755, "token_budget": 100000000}
     ]
    }
   ]
  },
  {
   "title": "Lagging test model (GPT-6 Luna) 100M budget",
   "x": {"label": "tokens", "unit": "", "scale": "linear"},
   "y": {"label": "", "unit": "", "scale": "linear"},
   "tables": [
    {
     "name": "token use",
     "rows": [
      {"model": "GPT-4o", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-4o", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "o1", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 206352, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 6124392, "token_budget": 100000000},
      {"model": "Claude Opus 4.5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 854669, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 161622, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 1786023, "token_budget": 100000000},
      {"model": "Claude Opus 4.6", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 39010, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 7576506, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 5878057, "token_budget": 100000000},
      {"model": "GPT-5.4", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 401504, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 3586475, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 3011201, "token_budget": 100000000},
      {"model": "Claude Opus 4.7", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 10285911, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "GPT-5.5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 6159353, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 8037219, "token_budget": 100000000},
      {"model": "Claude Opus 4.8", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 13198663, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 50022360, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 78603565, "token_budget": 100000000},
      {"model": "Claude Fable 5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 29870680, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 14964408, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 10608232, "token_budget": 100000000},
      {"model": "GPT-5.6 Sol", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 5618871, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 0, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 22718655, "token_budget": 100000000},
      {"model": "Claude Opus 5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 28499828, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 19242487, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 24863188, "token_budget": 100000000},
      {"model": "Claude Fable 5.1", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 14510641, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 13491280, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 3349175, "token_budget": 100000000},
      {"model": "GPT-6 Astra", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 7927722, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 1", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 39392710, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 2", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 52387906, "token_budget": 100000000},
      {"model": "Claude Opus 5.5", "run": "Run 3", "token_pool": "Lagging test model (GPT-6 Luna)", "tokens_used": 41000720, "token_budget": 100000000}
     ]
    }
   ]
  }
 ]
}
