{
  "canitoolcall_version": "0.1.0.dev0",
  "cells": [
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 4,
          "pass": 24,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 4,
          "pass": 24,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 25,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 64,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 3,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 3,
          "pass": 61,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 29,
        "pass": 35,
        "soft_pass": 0,
        "unsupported": 21
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "deepseek",
      "page": "cells/llamacpp/a25c9865/deepseek.html",
      "pass_rate": 0.546875,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 29
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 10,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 10,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 0,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 0,
          "pass": 11,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 12,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 0,
          "pass": 12,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 12,
        "soft_pass": 0,
        "unsupported": 73
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "deepseek",
      "page": "cells/ollama/7af39318/deepseek.html",
      "pass_rate": 1.0,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "pass",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 6,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 20,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 24,
          "pass": 11,
          "soft_pass": 23,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 4,
          "pass": 60,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 3,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 22,
          "pass": 13,
          "soft_pass": 29,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 26,
          "pass": 13,
          "soft_pass": 25,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 28,
        "pass": 13,
        "soft_pass": 23,
        "unsupported": 21
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "deepseek",
      "page": "cells/sglang/0.5.20/deepseek.html",
      "pass_rate": 0.203125,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 85
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "deepseek",
      "page": "cells/transformers/5.17.0/deepseek.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 18,
          "pass": 54,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 22,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 6,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 26,
          "pass": 43,
          "soft_pass": 9,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 83,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 3,
          "pass": 8,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 25,
          "pass": 60,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 29,
          "pass": 56,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 33,
        "pass": 43,
        "soft_pass": 9,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "deepseek",
      "page": "cells/vllm/0.30.0/deepseek.html",
      "pass_rate": 0.5058823529411764,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 3,
          "pass": 36,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 3,
          "pass": 36,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 2,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 4,
          "pass": 35,
          "soft_pass": 5,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 2,
          "pass": 46,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 35,
        "soft_pass": 5,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "gemma4",
      "page": "cells/llamacpp/a25c9865/gemma4.html",
      "pass_rate": 0.7291666666666666,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 8
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 6,
          "pass": 37,
          "soft_pass": 1,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 7,
        "pass": 40,
        "soft_pass": 1,
        "unsupported": 0
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "gemma4",
      "page": "cells/ollama/7af39318/gemma4.html",
      "pass_rate": 0.8333333333333334,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "fail",
      "stress_failures": 7
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 4,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 6,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 2,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 9,
          "pass": 30,
          "soft_pass": 5,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 10,
          "pass": 38,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 13,
        "pass": 30,
        "soft_pass": 5,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "gemma4",
      "page": "cells/sglang/0.5.20/gemma4.html",
      "pass_rate": 0.625,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 7,
          "pass": 30,
          "soft_pass": 7,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 2,
          "pass": 39,
          "soft_pass": 7,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 33,
        "soft_pass": 7,
        "unsupported": 0
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "gemma4",
      "page": "cells/transformers/5.17.0/gemma4.html",
      "pass_rate": 0.6875,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "fail",
      "stress_failures": 8
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 2,
          "pass": 39,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 12,
          "pass": 30,
          "soft_pass": 2,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 4,
          "pass": 42,
          "soft_pass": 2,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 16,
        "pass": 30,
        "soft_pass": 2,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "gemma4",
      "page": "cells/vllm/0.30.0/gemma4.html",
      "pass_rate": 0.625,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 6,
          "pass": 37,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 6,
          "pass": 37,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 3,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 5,
          "pass": 41,
          "soft_pass": 2,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 3,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 41,
        "soft_pass": 2,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "glm",
      "page": "cells/llamacpp/a25c9865/glm.html",
      "pass_rate": 0.803921568627451,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 8
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 27,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 27,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 2,
          "pass": 28,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 31,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 0,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 3,
        "pass": 29,
        "soft_pass": 0,
        "unsupported": 19
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "glm",
      "page": "cells/ollama/7af39318/glm.html",
      "pass_rate": 0.90625,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "fail",
      "stress_failures": 3
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 6,
          "pass": 38,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 6,
          "pass": 38,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 3,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 21,
          "pass": 25,
          "soft_pass": 2,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 12,
          "pass": 39,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 8,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 12,
          "pass": 32,
          "soft_pass": 7,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 23,
          "pass": 26,
          "soft_pass": 2,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 24,
        "pass": 25,
        "soft_pass": 2,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "glm",
      "page": "cells/sglang/0.5.20/glm.html",
      "pass_rate": 0.49019607843137253,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 51
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "glm",
      "page": "cells/transformers/5.17.0/glm.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 4,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 4,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 3,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 4,
          "pass": 33,
          "soft_pass": 11,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 8,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 50,
          "soft_pass": 1,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 5,
          "pass": 35,
          "soft_pass": 11,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 7,
        "pass": 33,
        "soft_pass": 11,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "glm",
      "page": "cells/vllm/0.30.0/glm.html",
      "pass_rate": 0.6470588235294118,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 1,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 1,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 5,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 3,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 51,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 2,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 43,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "gpt-oss",
      "page": "cells/llamacpp/a25c9865/gpt-oss.html",
      "pass_rate": 0.8431372549019608,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 8
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 5,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 4,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 51,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 4,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 6,
        "pass": 45,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "gpt-oss",
      "page": "cells/ollama/7af39318/gpt-oss.html",
      "pass_rate": 0.8823529411764706,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "fail",
      "stress_failures": 6
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 0,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 32,
          "pass": 13,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 29,
          "pass": 22,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 10,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 30,
          "pass": 21,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 32,
        "pass": 19,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "gpt-oss",
      "page": "cells/sglang/0.5.20/gpt-oss.html",
      "pass_rate": 0.37254901960784315,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 51
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "gpt-oss",
      "page": "cells/transformers/5.17.0/gpt-oss.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 2,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 2,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 3,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 1,
          "pass": 44,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 51,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 1,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 5,
        "pass": 46,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "gpt-oss",
      "page": "cells/vllm/0.30.0/gpt-oss.html",
      "pass_rate": 0.9019607843137255,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 5,
          "pass": 26,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 6,
          "pass": 25,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 2,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 11,
          "pass": 35,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 10,
          "pass": 39,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 14,
        "pass": 35,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "kimi",
      "page": "cells/llamacpp/a25c9865/kimi.html",
      "pass_rate": 0.7142857142857143,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 14
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 49
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "kimi",
      "page": "cells/ollama/7af39318/kimi.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 2,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 2,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 4,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 4,
          "pass": 45,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 5,
        "pass": 44,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "kimi",
      "page": "cells/sglang/0.5.20/kimi.html",
      "pass_rate": 0.8979591836734694,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 49
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "kimi",
      "page": "cells/transformers/5.17.0/kimi.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 2,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 2,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 7,
          "pass": 20,
          "soft_pass": 19,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 4,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 1,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 6,
          "pass": 43,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 22,
        "soft_pass": 19,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "kimi",
      "page": "cells/vllm/0.30.0/kimi.html",
      "pass_rate": 0.4489795918367347,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 5,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 5,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 9,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 24,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 4,
          "pass": 20,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 10,
        "pass": 14,
        "soft_pass": 0,
        "unsupported": 10
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "llama",
      "page": "cells/llamacpp/a25c9865/llama.html",
      "pass_rate": 0.5833333333333334,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 10
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 34
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "llama",
      "page": "cells/ollama/7af39318/llama.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 19,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 19,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 23,
          "pass": 9,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 1,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 19,
          "pass": 15,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 21,
          "pass": 13,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 25,
        "pass": 9,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "llama",
      "page": "cells/sglang/0.5.20/llama.html",
      "pass_rate": 0.2647058823529412,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 34
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "llama",
      "page": "cells/transformers/5.17.0/llama.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 3,
          "pass": 25,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 3,
          "pass": 25,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 2,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 23,
          "pass": 9,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 34,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 18,
          "pass": 16,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 25,
          "pass": 9,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 25,
        "pass": 9,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "llama",
      "page": "cells/vllm/0.30.0/llama.html",
      "pass_rate": 0.2647058823529412,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 12,
          "pass": 21,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 12,
          "pass": 21,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 2,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 17,
          "pass": 28,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 45,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 5,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 11,
          "pass": 36,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 19,
        "pass": 28,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "mistral",
      "page": "cells/llamacpp/a25c9865/mistral.html",
      "pass_rate": 0.5957446808510638,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 19
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 13,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 13,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 0,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 0,
          "pass": 14,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 14,
        "soft_pass": 0,
        "unsupported": 33
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "mistral",
      "page": "cells/ollama/7af39318/mistral.html",
      "pass_rate": 1.0,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "pass",
      "stress_failures": 2
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 7,
          "pass": 27,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 7,
          "pass": 27,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 0,
          "pass": 2,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 23,
          "pass": 21,
          "soft_pass": 1,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 5,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 8,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 20,
          "pass": 27,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 21,
          "pass": 26,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 23,
        "pass": 23,
        "soft_pass": 1,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "mistral",
      "page": "cells/sglang/0.5.20/mistral.html",
      "pass_rate": 0.48936170212765956,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 47
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "mistral",
      "page": "cells/transformers/5.17.0/mistral.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 6,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 6,
          "pass": 33,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 5,
          "pass": 38,
          "soft_pass": 2,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 4,
          "pass": 43,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 3,
          "pass": 44,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 6,
          "pass": 41,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 7,
        "pass": 38,
        "soft_pass": 2,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "mistral",
      "page": "cells/vllm/0.30.0/mistral.html",
      "pass_rate": 0.8085106382978723,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 4,
          "pass": 28,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 4,
          "pass": 28,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 6,
          "pass": 21,
          "soft_pass": 16,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 5,
          "pass": 45,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 1,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 11,
        "pass": 23,
        "soft_pass": 16,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "qwen3-hermes",
      "page": "cells/llamacpp/a25c9865/qwen3-hermes.html",
      "pass_rate": 0.46,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 11
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 1,
          "pass": 33,
          "soft_pass": 9,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 0,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 2,
        "pass": 39,
        "soft_pass": 9,
        "unsupported": 0
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "qwen3-hermes",
      "page": "cells/ollama/7af39318/qwen3-hermes.html",
      "pass_rate": 0.78,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "fail",
      "stress_failures": 2
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 32,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 32,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 33,
          "pass": 6,
          "soft_pass": 4,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 6,
          "pass": 44,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 7,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 33,
          "pass": 17,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 39,
          "pass": 11,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 39,
        "pass": 7,
        "soft_pass": 4,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "qwen3-hermes",
      "page": "cells/sglang/0.5.20/qwen3-hermes.html",
      "pass_rate": 0.14,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 50
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "qwen3-hermes",
      "page": "cells/transformers/5.17.0/qwen3-hermes.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 5,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 5,
          "pass": 32,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 4,
          "pass": 3,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 2,
          "pass": 23,
          "soft_pass": 18,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 2,
          "pass": 48,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 9,
          "pass": 33,
          "soft_pass": 8,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 9,
        "pass": 23,
        "soft_pass": 18,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "qwen3-hermes",
      "page": "cells/vllm/0.30.0/qwen3-hermes.html",
      "pass_rate": 0.46,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 5,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 5,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 6,
          "pass": 24,
          "soft_pass": 23,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 4,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 5,
          "pass": 49,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 7,
        "pass": 24,
        "soft_pass": 23,
        "unsupported": 0
      },
      "engine": "llamacpp",
      "engine_version": "a25c9865",
      "family": "qwen3-xml",
      "page": "cells/llamacpp/a25c9865/qwen3-xml.html",
      "pass_rate": 0.4444444444444444,
      "run_at": "2026-09-26T05:37:20Z",
      "status": "fail",
      "stress_failures": 7
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 0,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 0,
          "pass": 42,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 0,
          "pass": 1,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 9,
          "pass": 44,
          "soft_pass": 0,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 3,
          "pass": 51,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 2,
          "pass": 52,
          "soft_pass": 0,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 9,
        "pass": 45,
        "soft_pass": 0,
        "unsupported": 0
      },
      "engine": "ollama",
      "engine_version": "7af39318",
      "family": "qwen3-xml",
      "page": "cells/ollama/7af39318/qwen3-xml.html",
      "pass_rate": 0.8333333333333334,
      "run_at": "2026-09-26T05:36:50Z",
      "status": "fail",
      "stress_failures": 10
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 2,
          "pass": 47,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 3,
          "pass": 46,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 7,
          "pass": 19,
          "soft_pass": 27,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 4,
          "pass": 50,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 1,
          "pass": 6,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 54,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 8,
          "pass": 40,
          "soft_pass": 6,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 8,
        "pass": 19,
        "soft_pass": 27,
        "unsupported": 0
      },
      "engine": "sglang",
      "engine_version": "0.5.20",
      "family": "qwen3-xml",
      "page": "cells/sglang/0.5.20/qwen3-xml.html",
      "pass_rate": 0.35185185185185186,
      "run_at": "2026-09-26T05:22:38Z",
      "status": "fail",
      "stress_failures": 0
    },
    {
      "checks": {},
      "counts": {
        "error": 0,
        "fail": 0,
        "pass": 0,
        "soft_pass": 0,
        "unsupported": 54
      },
      "engine": "transformers",
      "engine_version": "5.17.0",
      "family": "qwen3-xml",
      "page": "cells/transformers/5.17.0/qwen3-xml.html",
      "pass_rate": null,
      "run_at": "2026-09-26T05:18:47Z",
      "status": "unsupported",
      "stress_failures": 0
    },
    {
      "checks": {
        "arguments_json": {
          "error": 0,
          "fail": 3,
          "pass": 46,
          "soft_pass": 0,
          "unsupported": 0
        },
        "arguments_schema": {
          "error": 0,
          "fail": 3,
          "pass": 46,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_error": {
          "error": 0,
          "fail": 1,
          "pass": 0,
          "soft_pass": 0,
          "unsupported": 0
        },
        "expected_match": {
          "error": 0,
          "fail": 4,
          "pass": 24,
          "soft_pass": 25,
          "unsupported": 0
        },
        "no_leakage": {
          "error": 0,
          "fail": 0,
          "pass": 54,
          "soft_pass": 0,
          "unsupported": 0
        },
        "parallel_order": {
          "error": 0,
          "fail": 0,
          "pass": 7,
          "soft_pass": 0,
          "unsupported": 0
        },
        "split_invariance": {
          "error": 0,
          "fail": 0,
          "pass": 54,
          "soft_pass": 0,
          "unsupported": 0
        },
        "stream_equals_nonstream": {
          "error": 0,
          "fail": 4,
          "pass": 45,
          "soft_pass": 5,
          "unsupported": 0
        }
      },
      "counts": {
        "error": 0,
        "fail": 5,
        "pass": 24,
        "soft_pass": 25,
        "unsupported": 0
      },
      "engine": "vllm",
      "engine_version": "0.30.0",
      "family": "qwen3-xml",
      "page": "cells/vllm/0.30.0/qwen3-xml.html",
      "pass_rate": 0.4444444444444444,
      "run_at": "2026-09-26T05:20:52Z",
      "status": "fail",
      "stress_failures": 0
    }
  ],
  "engines": [
    {
      "commit": "a25c9865fe03c954c93fd755b5d79ae86ba99750",
      "name": "llamacpp",
      "run_finished_at": "2026-09-26T05:37:20Z",
      "run_started_at": "2026-09-26T05:36:17Z",
      "version": "a25c9865"
    },
    {
      "commit": "7af393188defd52d370464de0d2064649cab9b41",
      "name": "ollama",
      "run_finished_at": "2026-09-26T05:36:50Z",
      "run_started_at": "2026-09-26T05:36:43Z",
      "version": "7af39318"
    },
    {
      "commit": null,
      "name": "sglang",
      "run_finished_at": "2026-09-26T05:22:38Z",
      "run_started_at": "2026-09-26T05:19:05Z",
      "version": "0.5.20"
    },
    {
      "commit": null,
      "name": "transformers",
      "run_finished_at": "2026-09-26T05:18:47Z",
      "run_started_at": "2026-09-26T05:18:44Z",
      "version": "5.17.0"
    },
    {
      "commit": null,
      "name": "vllm",
      "run_finished_at": "2026-09-26T05:20:52Z",
      "run_started_at": "2026-09-26T05:19:00Z",
      "version": "0.30.0"
    }
  ],
  "families": [
    "deepseek",
    "gemma4",
    "glm",
    "gpt-oss",
    "kimi",
    "llama",
    "mistral",
    "qwen3-hermes",
    "qwen3-xml"
  ]
}
