{
  "base_url": "http://127.0.0.1:8001/v1",
  "budget": {
    "max_completion_tokens": 8192,
    "reasoning_headroom_tokens": 7168,
    "visible_answer_tokens": 1024
  },
  "checks": [
    "smoke",
    "json",
    "needle",
    "tools",
    "long-tools",
    "streaming-tools",
    "tool-result",
    "responses"
  ],
  "evidence_policy": {
    "allowed_finish_reasons": [
      "stop",
      "tool_calls"
    ],
    "errors": [],
    "reasoning": "any"
  },
  "image_expect": [],
  "model": "glm53-flash-exl3-k3-dflash2-k5-fp8-tp2-524k-vision-xgfix",
  "multimodal_input": null,
  "observations": [
    {
      "content": "```python\nsum(xs)\n```\n\nThat's it \u2014 Python's built-in `sum()` does exactly this. If you want it as a lambda:\n\n```python\ntotal = lambda xs: sum(xs)\n```",
      "content_chars": 149,
      "content_excerpt": "```python\nsum(xs)\n```\n\nThat's it \u2014 Python's built-in `sum()` does exactly this. If you want it as a lambda:\n\n```python\ntotal = lambda xs: sum(xs)\n```",
      "finish_reason": "stop",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "seconds": 1.426,
      "test": "smoke",
      "usage": {
        "completion_tokens": 45,
        "prompt_tokens": 28,
        "prompt_tokens_details": null,
        "total_tokens": 73
      },
      "validation_detail": "contains sum("
    },
    {
      "content": "{\"language\":\"python\",\"ok\":true}",
      "content_chars": 31,
      "content_excerpt": "{\"language\":\"python\",\"ok\":true}",
      "finish_reason": "stop",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "seconds": 0.174,
      "test": "json",
      "usage": {
        "completion_tokens": 11,
        "prompt_tokens": 30,
        "prompt_tokens_details": null,
        "total_tokens": 41
      },
      "validation_detail": "parsed keys=['language', 'ok']"
    },
    {
      "content": "ZEBRA-42917-QUARTZ",
      "content_chars": 18,
      "content_excerpt": "ZEBRA-42917-QUARTZ",
      "finish_reason": "stop",
      "passed": true,
      "reasoning_chars": 18,
      "reasoning_excerpt": "ZEBRA-42917-QUARTZ",
      "reasoning_field": "reasoning",
      "reasoning_tokens": null,
      "seconds": 67.025,
      "test": "needle",
      "usage": {
        "completion_tokens": 24,
        "prompt_tokens": 206296,
        "prompt_tokens_details": null,
        "total_tokens": 206320
      },
      "validation_detail": "needle present"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 2,
      "seconds": 3.715,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 0,
      "seconds": 5.445,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 1,
      "seconds": 6.614,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 3,
      "seconds": 8.327,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 5,
      "seconds": 9.49,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 6,
      "seconds": 11.276,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 4,
      "seconds": 12.479,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 7,
      "seconds": 13.667,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 8,
      "seconds": 15.417,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 9,
      "seconds": 16.597,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 10,
      "seconds": 18.335,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 11,
      "seconds": 19.506,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 13,
      "seconds": 21.231,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 12,
      "seconds": 22.406,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 14,
      "seconds": 24.133,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 15,
      "seconds": 25.303,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 16,
      "seconds": 27.639,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 17,
      "seconds": 28.212,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 19,
      "seconds": 33.863,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "request_index": 18,
      "seconds": 33.973,
      "test": "tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 4967,
        "prompt_tokens_details": null,
        "total_tokens": 4980
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "measured_prompt_tokens": 211101,
      "minimum_prompt_tokens": 100000,
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "requested_context_tokens": 250000,
      "seconds": 62.556,
      "test": "long-tools",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 211101,
        "prompt_tokens_details": null,
        "total_tokens": 211114
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content_chars": 0,
      "content_excerpt": "",
      "event_count": 13,
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 123,
      "reasoning_excerpt": "The user wants to know the weather in Oakland. I have a get_weather tool available. Let me call it with the city \"Oakland\".",
      "reasoning_field": "stream_delta",
      "reasoning_tokens": null,
      "seconds": 5.537,
      "sse_done": true,
      "test": "streaming-tools",
      "usage": {
        "completion_tokens": 42,
        "prompt_tokens": 166,
        "total_tokens": 208
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "",
      "content_chars": 0,
      "content_excerpt": "",
      "finish_reason": "tool_calls",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "seconds": 0.255,
      "test": "tool-result-initial",
      "usage": {
        "completion_tokens": 13,
        "prompt_tokens": 166,
        "prompt_tokens_details": null,
        "total_tokens": 179
      },
      "validation_detail": "valid tool_call get_weather(city='Oakland')"
    },
    {
      "content": "OAKLAND 72F",
      "content_chars": 11,
      "content_excerpt": "OAKLAND 72F",
      "finish_reason": "stop",
      "passed": true,
      "reasoning_chars": 0,
      "reasoning_excerpt": "",
      "reasoning_field": null,
      "reasoning_tokens": null,
      "seconds": 0.256,
      "test": "tool-result-continuation",
      "usage": {
        "completion_tokens": 8,
        "prompt_tokens": 70,
        "prompt_tokens_details": null,
        "total_tokens": 78
      },
      "validation_detail": "tool result retained"
    },
    {
      "content_chars": 5,
      "content_excerpt": "READY",
      "finish_reason": "stop",
      "passed": true,
      "reasoning_chars": 135,
      "reasoning_excerpt": "The user wants me to reply with exactly \"READY\". This is a simple instruction. I should just output the word \"READY\" with nothing else.",
      "reasoning_field": "reasoning_text",
      "reasoning_tokens": null,
      "response_status": "completed",
      "seconds": 0.478,
      "test": "responses",
      "usage": {
        "input_tokens": 16,
        "input_tokens_details": {
          "cached_tokens": 0,
          "cached_tokens_per_turn": [],
          "input_tokens_per_turn": []
        },
        "output_tokens": 33,
        "output_tokens_details": {
          "output_tokens_per_turn": [],
          "reasoning_tokens": 0,
          "tool_output_tokens": 0,
          "tool_output_tokens_per_turn": []
        },
        "total_tokens": 49
      },
      "validation_detail": "completed exact READY"
    }
  ],
  "ocr_expect": [],
  "passed": true,
  "results": [
    {
      "detail": "1.4s got=\"```python\\nsum(xs)\\n```\\n\\nThat's it \u2014 Python's built-\" finish='stop' visible=149 reasoning_chars=0 reasoning_tokens=None",
      "name": "smoke (short coding)",
      "passed": true
    },
    {
      "detail": "parsed keys=['language', 'ok'] finish='stop' visible=31 reasoning_chars=0 reasoning_tokens=None",
      "name": "structured JSON",
      "passed": true
    },
    {
      "detail": "67.0s ctx~250000 got='ZEBRA-42917-QUARTZ' finish='stop' visible=18 reasoning_chars=18 reasoning_tokens=None",
      "name": "needle @ ~250000 ctx",
      "passed": true
    },
    {
      "detail": "20/20 clean (sample: 3.7s valid tool_call get_weather(city='Oakland') finish='tool_calls' visible=0 reasoning_chars=0 reasoning_tokens=None)",
      "name": "shared-prefix tool batch x20",
      "passed": true
    },
    {
      "detail": "62.6s measured_prompt=211101 valid tool_call get_weather(city='Oakland') finish='tool_calls' visible=0 reasoning_chars=0 reasoning_tokens=None",
      "name": "tool call after >=100K prompt (target 250000)",
      "passed": true
    },
    {
      "detail": "5.5s done=True events=13 valid tool_call get_weather(city='Oakland') finish='tool_calls'",
      "name": "streaming tool call",
      "passed": true
    },
    {
      "detail": "0.5s tool result retained finish='stop' visible=11 reasoning_chars=0 reasoning_tokens=None",
      "name": "tool-result continuation",
      "passed": true
    },
    {
      "detail": "0.5s status='completed' output='READY' reasoning_chars=135",
      "name": "Responses API subset",
      "passed": true
    }
  ],
  "schema_version": "preflight/v2",
  "thinking": {
    "chat_template_kwargs": null,
    "mode": "default",
    "reasoning_effort": "low"
  },
  "video_expect": [],
  "video_input": null
}
