{
  "schema_version": 2,
  "suite_id": "infra-core",
  "arenas_requested": 4,
  "arenas_completed": 4,
  "plan": [
    {
      "arena_id": "first-build-real",
      "manifest": "suites/../arenas/first-build/agents-real.toml",
      "compatibility_key": "a0c9ad7f3518991b3f5a16832777737070f98f48d2b32cb223b1ebb89cf3ec15",
      "rounds": 1,
      "output": "benchmarks/infra-core-smoke-20260815-003451/01-first-build-real"
    },
    {
      "arena_id": "durable-job-queue",
      "manifest": "suites/../arenas/durable-job-queue/agents-real.toml",
      "compatibility_key": "a6420c177333e37325ef991c9d9ac83f34c03a1fdf6dd9f67b9881696d95b9c6",
      "rounds": 1,
      "output": "benchmarks/infra-core-smoke-20260815-003451/02-durable-job-queue"
    },
    {
      "arena_id": "zero-downtime-rollout",
      "manifest": "suites/../arenas/zero-downtime-rollout/agents-real.toml",
      "compatibility_key": "9ff8fcd18e643079e3aaf0a725cbb847da8351cf4c732ce2b454e5e4f48bc489",
      "rounds": 1,
      "output": "benchmarks/infra-core-smoke-20260815-003451/03-zero-downtime-rollout"
    },
    {
      "arena_id": "primary-failover",
      "manifest": "suites/../arenas/primary-failover/agents-real.toml",
      "compatibility_key": "8e3fdce8d53fcf1f29e4a1f63fa8f8ef8b6ea81ab305983763318ec900188d18",
      "rounds": 1,
      "output": "benchmarks/infra-core-smoke-20260815-003451/04-primary-failover"
    }
  ],
  "arenas": [
    {
      "arena_id": "first-build-real",
      "output": "benchmarks/infra-core-smoke-20260815-003451/01-first-build-real",
      "rounds_requested": 1,
      "rounds_completed": 1,
      "aborted": false,
      "standings": [
        {
          "model": "openai/gpt-5.6-luna",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 1,
          "durable_deployments": 1,
          "durable_times_ms": [
            44431
          ],
          "median_durable_ms": 44431,
          "milestone_passes": 4,
          "milestones_available": 4,
          "usage_recorded": 1,
          "input_tokens": 27669,
          "output_tokens": 2019,
          "cost_microusd": 2266,
          "cost_per_durable_microusd": 2266,
          "failures": {
            "arena": 1
          }
        },
        {
          "model": "z-ai/glm-5.2",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 2,
          "milestones_available": 4,
          "usage_recorded": 1,
          "input_tokens": 39732,
          "output_tokens": 2243,
          "cost_microusd": 17262,
          "cost_per_durable_microusd": null,
          "failures": {
            "player": 1
          }
        },
        {
          "model": "deepseek/deepseek-v4-flash-0731",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 0,
          "milestones_available": 4,
          "usage_recorded": 1,
          "input_tokens": 33545,
          "output_tokens": 2197,
          "cost_microusd": 1424,
          "cost_per_durable_microusd": null,
          "failures": {
            "player": 1
          }
        }
      ]
    },
    {
      "arena_id": "durable-job-queue",
      "output": "benchmarks/infra-core-smoke-20260815-003451/02-durable-job-queue",
      "rounds_requested": 1,
      "rounds_completed": 1,
      "aborted": false,
      "standings": [
        {
          "model": "openai/gpt-5.6-luna",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 1,
          "durable_deployments": 1,
          "durable_times_ms": [
            112880
          ],
          "median_durable_ms": 112880,
          "milestone_passes": 5,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 106279,
          "output_tokens": 6518,
          "cost_microusd": 6388,
          "cost_per_durable_microusd": 6388,
          "failures": {
            "arena": 1
          }
        },
        {
          "model": "z-ai/glm-5.2",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 1,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 194015,
          "output_tokens": 10158,
          "cost_microusd": 91563,
          "cost_per_durable_microusd": null,
          "failures": {
            "player": 1
          }
        },
        {
          "model": "deepseek/deepseek-v4-flash-0731",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 0,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 63365,
          "output_tokens": 4484,
          "cost_microusd": 2501,
          "cost_per_durable_microusd": null,
          "failures": {
            "controller": 1
          }
        }
      ]
    },
    {
      "arena_id": "zero-downtime-rollout",
      "output": "benchmarks/infra-core-smoke-20260815-003451/03-zero-downtime-rollout",
      "rounds_requested": 1,
      "rounds_completed": 1,
      "aborted": false,
      "standings": [
        {
          "model": "openai/gpt-5.6-luna",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 1,
          "durable_deployments": 1,
          "durable_times_ms": [
            79125
          ],
          "median_durable_ms": 79125,
          "milestone_passes": 5,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 135896,
          "output_tokens": 5236,
          "cost_microusd": 6335,
          "cost_per_durable_microusd": 6335,
          "failures": {
            "arena": 1
          }
        },
        {
          "model": "deepseek/deepseek-v4-flash-0731",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 1,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 87843,
          "output_tokens": 2841,
          "cost_microusd": 2591,
          "cost_per_durable_microusd": null,
          "failures": {
            "player": 1
          }
        },
        {
          "model": "z-ai/glm-5.2",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 1,
          "milestones_available": 5,
          "usage_recorded": 1,
          "input_tokens": 76193,
          "output_tokens": 3044,
          "cost_microusd": 17467,
          "cost_per_durable_microusd": null,
          "failures": {
            "player": 1
          }
        }
      ]
    },
    {
      "arena_id": "primary-failover",
      "output": "benchmarks/infra-core-smoke-20260815-003451/04-primary-failover",
      "rounds_requested": 1,
      "rounds_completed": 1,
      "aborted": false,
      "standings": [
        {
          "model": "openai/gpt-5.6-luna",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 1,
          "durable_deployments": 1,
          "durable_times_ms": [
            66308
          ],
          "median_durable_ms": 66308,
          "milestone_passes": 6,
          "milestones_available": 6,
          "usage_recorded": 1,
          "input_tokens": 45595,
          "output_tokens": 3194,
          "cost_microusd": 3638,
          "cost_per_durable_microusd": 3638,
          "failures": {
            "arena": 1
          }
        },
        {
          "model": "deepseek/deepseek-v4-flash-0731",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 1,
          "milestones_available": 6,
          "usage_recorded": 1,
          "input_tokens": 28998,
          "output_tokens": 2529,
          "cost_microusd": 1333,
          "cost_per_durable_microusd": null,
          "failures": {
            "controller": 1
          }
        },
        {
          "model": "z-ai/glm-5.2",
          "adapter": "claux",
          "reasoning_effort": "high",
          "appearances": 1,
          "wins": 0,
          "durable_deployments": 0,
          "durable_times_ms": [],
          "median_durable_ms": null,
          "milestone_passes": 1,
          "milestones_available": 6,
          "usage_recorded": 1,
          "input_tokens": 69456,
          "output_tokens": 4691,
          "cost_microusd": 17161,
          "cost_per_durable_microusd": null,
          "failures": {
            "controller": 1
          }
        }
      ]
    }
  ],
  "standings": [
    {
      "model": "openai/gpt-5.6-luna",
      "adapter": "claux",
      "reasoning_effort": "high",
      "appearances": 4,
      "wins": 4,
      "durable_deployments": 4,
      "durable_times_ms": [
        44431,
        66308,
        79125,
        112880
      ],
      "median_durable_ms": 79125,
      "milestone_passes": 20,
      "milestones_available": 20,
      "usage_recorded": 4,
      "input_tokens": 315439,
      "output_tokens": 16967,
      "cost_microusd": 18627,
      "cost_per_durable_microusd": 4656,
      "failures": {
        "arena": 4
      }
    },
    {
      "model": "z-ai/glm-5.2",
      "adapter": "claux",
      "reasoning_effort": "high",
      "appearances": 4,
      "wins": 0,
      "durable_deployments": 0,
      "durable_times_ms": [],
      "median_durable_ms": null,
      "milestone_passes": 5,
      "milestones_available": 20,
      "usage_recorded": 4,
      "input_tokens": 379396,
      "output_tokens": 20136,
      "cost_microusd": 143453,
      "cost_per_durable_microusd": null,
      "failures": {
        "controller": 1,
        "player": 3
      }
    },
    {
      "model": "deepseek/deepseek-v4-flash-0731",
      "adapter": "claux",
      "reasoning_effort": "high",
      "appearances": 4,
      "wins": 0,
      "durable_deployments": 0,
      "durable_times_ms": [],
      "median_durable_ms": null,
      "milestone_passes": 2,
      "milestones_available": 20,
      "usage_recorded": 4,
      "input_tokens": 213751,
      "output_tokens": 12051,
      "cost_microusd": 7849,
      "cost_per_durable_microusd": null,
      "failures": {
        "controller": 2,
        "player": 2
      }
    }
  ]
}