{
  "profile": "Chat Completions only; six requests maximum; no retries",
  "model_requested": "openai/gpt-6-luna",
  "endpoint": "https://api.ofox.ai/v1/chat/completions",
  "requests": 6,
  "results": [
    {
      "check": "chat",
      "status": "PASS",
      "detail": {
        "model_returned": "openai/gpt-6-luna",
        "finish_reason": "stop",
        "usage": {
          "prompt_tokens": 10,
          "completion_tokens": 4,
          "total_tokens": 14
        }
      },
      "elapsed_seconds": 2.147
    },
    {
      "check": "structured-output",
      "status": "PASS",
      "detail": {
        "schema_and_values_checked": true,
        "usage": {
          "prompt_tokens": 54,
          "completion_tokens": 33,
          "total_tokens": 87,
          "completion_tokens_details": {
            "reasoning_tokens": 11
          }
        }
      },
      "elapsed_seconds": 2.065
    },
    {
      "check": "tool-round-trip",
      "status": "PASS",
      "detail": {
        "tool_round_trip": true,
        "first_usage": {
          "prompt_tokens": 72,
          "completion_tokens": 19,
          "total_tokens": 91
        },
        "final_usage": {
          "prompt_tokens": 115,
          "completion_tokens": 33,
          "total_tokens": 148,
          "completion_tokens_details": {
            "reasoning_tokens": 26
          }
        },
        "review_note": "Final answer must equal 17 exactly; fixed synthetic task only.",
        "final_answer": "17"
      },
      "elapsed_seconds": 3.853
    },
    {
      "check": "streaming",
      "status": "PASS",
      "detail": {
        "json_events": 3,
        "finish_reason": "stop",
        "done": true,
        "usage": {
          "prompt_tokens": 10,
          "prompt_tokens_details": {
            "cached_tokens": 0,
            "cache_write_tokens": 0
          },
          "completion_tokens": 4,
          "completion_tokens_details": {},
          "total_tokens": 14
        },
        "note": "No stream_options usage request: absent streamed usage is recorded, not treated as failure."
      },
      "elapsed_seconds": 2.037
    },
    {
      "check": "error-shape",
      "status": "PASS",
      "detail": {
        "http_status": 400,
        "error_keys": [
          "code",
          "message",
          "request_id",
          "type"
        ]
      },
      "elapsed_seconds": 0.912
    }
  ],
  "limits": "No Responses, streaming tool arguments, load, fallback, all schemas, pricing or security certification tested. Output cap is requested on every call; provider support must be checked. Set account spend limits; a token parameter is not a guaranteed dollar cap. Socket inactivity timeout 45s; a best-effort stream deadline is checked between complete lines. This is not a wall-clock deadline: a trickling response can run longer. Interrupt a stalled run; do not deploy this starter as an unattended worker."
}