文档索引
在以下地址获取完整的文档索引：https://docs.typesafe.ai/llms.txt 使用此文件可以在进一步探索之前发现所有可用页面。

回答与响应

读取 TypeSafe API 返回的回答、置信度评分、token 使用情况和可用模型。

Response

typesafe_sdk.SystemOneResponse

pydantic-model

基类：Response

按问题类型分组的回答，包含模型和使用元数据。

请参阅 System One 了解详情。

Note

显示 JSON 架构：

{
  "$defs": {
    "ChoiceAnswer": {
      "description": "选定的标签及其概率。\n\n有关详细信息，请参阅 [choice 原语](https://docs.typesafe.ai/primitives/choice)。",
      "properties": {
        "type": {
          "const": "choice",
          "default": "choice",
          "title": "类型",
          "type": "string"
        },
        "choice": {
          "description": "在问题的条件中概率最高的选项名称。",
          "examples": [
            "angry"
          ],
          "title": "选项",
          "type": "string"
        },
        "confidence": {
          "description": "对选定选项的置信度，范围从 0 到 1。较高的值表示更大的确定性；使用较低的值来标记不确定的选择以供审核。",
          "examples": [
            0.9
          ],
          "title": "置信度",
          "type": "number"
        },
        "probabilities": {
          "additionalProperties": {
            "type": "number"
          },
          "description": "条件中每个选项的概率，以选项名称为键，范围从 0 到 1。显示替代方案的可能性；值的总和约为 1。",
          "examples": [
            {
              "angry": 0.8,
              "calm": 0.1,
              "excited": 0.1
            }
          ],
          "title": "概率",
          "type": "object"
        }
      },
      "required": [
        "choice",
        "confidence",
        "probabilities"
      ],
      "title": "ChoiceAnswer",
      "type": "object"
    },
    "NoulAnswer": {
      "description": "是/否答案。\n\n有关详细信息，请参阅 [noul 原语](https://docs.typesafe.ai/primitives/noul)。",
      "properties": {
        "type": {
          "const": "noul",
          "default": "noul",
          "title": "类型",
          "type": "string"
        },
        "noul": {
          "description": "是答案或真陈述的概率，范围从 0 到 1。接近 1 的值倾向于是或真，接近 0 的值倾向于否或假，接近 0.5 的值表示不确定。",
          "examples": [
            0.98
          ],
          "title": "Noul",
          "type": "number"
        }
      },
      "required": [
        "noul"
      ],
      "title": "NoulAnswer",
      "type": "object"
    },
    "ScoreAnswer": {
      "description": "预期分数及其评分标准和概率。\n\n有关详细信息，请参阅 [score 原语](https://docs.typesafe.ai/primitives/score)。",
      "properties": {
        "type": {
          "const": "score",
          "default": "score",
          "title": "类型",
          "type": "string"
        },
        "score": {
          "description": "预期分数：评分标准级别的概率加权平均值。可能介于整数级别之间。",
          "examples": [
            1.7
          ],
          "title": "分数",
          "type": "number"
        },
        "confidence": {
          "description": "对分数的置信度，范围从 0 到 1。较高的值表示更大的确定性；使用较低的值来标记不确定的评分以供审核。",
          "examples": [
            0.9
          ],
          "title": "置信度",
          "type": "number"
        },
        "legend": {
          "additionalProperties": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "additionalProperties": true,
                "type": "object"
              },
              {
                "items": {},
                "type": "array"
              }
            ]
          },
          "title": "图例",
          "type": "object"
        },
        "probabilities": {
          "additionalProperties": {
            "type": "number"
          },
          "title": "概率",
          "type": "object"
        }
      },
      "required": [
        "score",
        "confidence",
        "legend",
        "probabilities"
      ],
      "title": "ScoreAnswer",
      "type": "object"
    },
    "Usage": {
      "description": "API 报告时，请求的令牌计数。",
      "properties": {
        "input_tokens": {
          "anyOf": [
            {
              "type": "integer"
            },
            {
              "type": "null"
            }
          ],
          "default": null,
          "title": "输入令牌"
        },
        "output_tokens": {
          "anyOf": [
            {
              "type": "integer"
            },
            {
              "type": "null"
            }
          ],
          "default": null,
          "title": "输出令牌"
        }
      },
      "title": "Usage",
      "type": "object"
    }
  },
  "description": "按问题类型分组的回答，以及模型和使用元数据。\n\n有关详细信息，请参阅 [System One](https://docs.typesafe.ai/concepts/system-one)。",
  "properties": {
    "model": {
      "title": "模型",
      "type": "string"
    },
    "usage": {
      "$ref": "#/$defs/Usage"
    },
    "answers": {
      "additionalProperties": {
        "discriminator": {
          "mapping": {
            "choice": "#/$defs/ChoiceAnswer",
            "noul": "#/$defs/NoulAnswer",
            "score": "#/$defs/ScoreAnswer"
          },
          "propertyName": "type"
        },
        "oneOf": [
          {
            "$ref": "#/$defs/NoulAnswer"
          },
          {
            "$ref": "#/$defs/ChoiceAnswer"
          },
          {
            "$ref": "#/$defs/ScoreAnswer"
          }
        ]
      },
      "title": "回答",
      "type": "object"
    }
  },
  "required": [
    "model",
    "usage"
  ],
  "title": "SystemOneResponse",
  "type": "object"
}

配置：

extra: ignore
frozen: True
strict: True

字段：

model (str)
usage (Usage)
answers (dict[str, Answer])

request_id

cached property

request_id: str

x-typesafe-request-id 响应头。

raw_http_response

property

raw_http_response: httpx2.Response

底层的 httpx2.Response，暴露状态、标头和正文。

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

model

pydantic-field

model: str

用于回答请求的模型。

usage

pydantic-field

usage: 使用情况

请求的 Token 使用情况。

answers

pydantic-field

answers: dict[str, Answer]

以问题名称为键的所有回答对象。

nouls

cached property

nouls: dict[str, NoulAnswer]

以问题名称为键的是/否答案。

choices

cached property

choices: dict[str, ChoiceAnswer]

以问题名称为键的 Choice 答案。

scores

cached property

scores: dict[str, ScoreAnswer]

以问题名称为键的 Score 回答。

typesafe_sdk.Usage

pydantic-model

基类：wire.Usage

API 报告时，请求的 token 计数。

Note

显示 JSON 架构：

{
  "description": "API 报告请求的令牌数量。",
  "properties": {
    "input_tokens": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "输入令牌"
    },
    "output_tokens": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "输出令牌"
    }
  },
  "title": "使用情况",
  "type": "object"
}

配置：

extra: ignore
frozen: True
strict: True

字段：

input_tokens (int | None)
output_tokens (int | None)

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

input_tokens

pydantic-field

input_tokens: int | None = None

使用的输入 token 数量，如果 API 未报告则为 None。

output_tokens

pydantic-field

output_tokens: int | None = None

使用的输出 Token 数量，如果 API 未报告则为 None。

回答

typesafe_sdk.NoulAnswer

pydantic-model

基类：wire.NoulAnswer

一个“是/否”答案。

详见 noul 原语。

Note

显示 JSON 架构：

{
  "description": "是/否答案。\n\n详见 [noul 原语](https://docs.typesafe.ai/primitives/noul)。",
  "properties": {
    "type": {
      "const": "noul",
      "default": "noul",
      "title": "类型",
      "type": "字符串"
    },
    "noul": {
      "description": "“是”答案或“真”陈述的概率，范围从 0 到 1。接近 1 的值倾向于“是”或“真”，接近 0 的值倾向于“否”或“假”，接近 0.5 的值表示不确定。",
      "examples": [
        0.98
      ],
      "title": "Noul",
      "type": "数字"
    }
  },
  "required": [
    "noul"
  ],
  "title": "NoulAnswer",
  "type": "对象"
}

配置：

extra: ignore
frozen: True
strict: True

字段：

<a href="/sdk/python/api/types/responses#typesafe_sdk.NoulAnswer.noul">noul</a> (<a href="https://docs.python.org/3/builtins/functions.html#float">float</a>)
type (<a href="https://docs.python.org/3/library/typing.html#typing.Literal">Literal</a>['noul'])

noul

pydantic-field

noul: float

回答“是”或陈述为真的概率，范围为 0 到 1。接近 1 的值倾向于“是”或“真”，接近 0 的值倾向于“否”或“假”，而接近 0.5 的值表示不确定性。

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

typesafe_sdk.ChoiceAnswer

pydantic-model

基类：wire.ChoiceAnswer

选定的标签及其概率。

详见 choice 原语。

Note

显示 JSON 架构：

{
  "description": "所选标签及其概率。\n\n详见[选择基元](https://docs.typesafe.ai/primitives/choice)。",
  "properties": {
    "type": {
      "const": "choice",
      "default": "choice",
      "title": "类型",
      "type": "string"
    },
    "choice": {
      "description": "问题标准中概率最高的选项名称。",
      "examples": [
        "angry"
      ],
      "title": "选项",
      "type": "string"
    },
    "confidence": {
      "description": "对所选选项的信心，范围为 0 到 1。值越高表示确定性越大；请使用较低的值来标记需要审核的不确定选择。",
      "examples": [
        0.9
      ],
      "title": "置信度",
      "type": "number"
    },
    "probabilities": {
      "additionalProperties": {
        "type": "number"
      },
      "description": "标准中每个选项的概率，以选项名称为键，范围为 0 到 1。显示替代方案的可能性；值之和约为 1。",
      "examples": [
        {
          "angry": 0.8,
          "calm": 0.1,
          "excited": 0.1
        }
      ],
      "title": "概率",
      "type": "object"
    }
  },
  "required": [
    "choice",
    "confidence",
    "probabilities"
  ],
  "title": "ChoiceAnswer",
  "type": "object"
}

配置：

extra: ignore
frozen: True
strict: True

字段：

choice (str)
confidence (float)
probabilities (dict[str, float])
type (Literal['choice'])

choice

pydantic-field

choice: str

问题标准中概率最高的选项名称。

confidence

pydantic-field

confidence: float

所选选项的置信度，范围从 0 到 1。值越高表示确定性越强；使用较低值将不确定的选择标记为待审核。

probabilities

pydantic-field

probabilities: dict[str, float]

标准中每个选择的选择概率，以选择名称为键，范围从 0 到 1。显示各替代选项的可能性；值之和约为 1。

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

typesafe_sdk.ScoreAnswer

pydantic-model

基类：wire.ScoreAnswer

预期分数及其评分标准和概率。

详见 score 原语。

Note

显示 JSON 架构：

{
  "description": "预期分数及其评分标准和概率。\n\n有关详细信息，请参阅 [评分原语 (score primitive)](https://docs.typesafe.ai/primitives/score)。",
  "properties": {
    "type": {
      "const": "score",
      "default": "score",
      "title": "类型 (Type)",
      "type": "string"
    },
    "score": {
      "description": "预期分数：评分标准层的概率加权平均值。可能介于整数层级之间。",
      "examples": [
        1.7
      ],
      "title": "分数 (Score)",
      "type": "number"
    },
    "confidence": {
      "description": "分数的置信度，范围从 0 到 1。值越高表示确定性越强；使用较低值将不确定的评级标记为待审核。",
      "examples": [
        0.9
      ],
      "title": "置信度 (Confidence)",
      "type": "number"
    },
    "legend": {
      "additionalProperties": {
        "anyOf": [
          {
            "type": "string"
          },
          {
            "additionalProperties": true,
            "type": "object"
          },
          {
            "items": {},
            "type": "array"
          }
        ]
      },
      "title": "图表/图例 (Legend)",
      "type": "object"
    },
    "probabilities": {
      "additionalProperties": {
        "type": "number"
      },
      "title": "概率 (Probabilities)",
      "type": "object"
    }
  },
  "required": [
    "score",
    "confidence",
    "legend",
    "probabilities"
  ],
  "title": "ScoreAnswer",
  "type": "object"
}

配置：

extra: ignore
frozen: True
strict: True

字段：

score (float)
confidence (float)
type (Literal['score'])
legend (dict[int, str | dict[str, Any] | list[Any]])
probabilities (dict[int, float])

score

pydantic-field

score: float

预期分数：评分标准级别的概率加权平均值。可能介于整数级别之间。

confidence

pydantic-field

confidence: float

对分数的置信度，范围为 0 到 1。值越高表示确定性越大；请使用较低的值来标记需要审核的不确定评分。

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

legend

pydantic-field

legend: dict[
    int, str | dict[str, Any] | list[Any]
]

以整数分数为键的评分标准描述。

probabilities

pydantic-field

probabilities: dict[int, float]

以整数分数为键的概率。

typesafe_sdk.Answer

module-attribute

回答: 类型别名 = 已注解[
    是否回答 | 选择回答 | 评分回答,
    字段(鉴别器="类型"),
]

对单个问题的回答，通过其 type 标识。

可用模型

typesafe_sdk.ListModelsResponse

pydantic-model

基类：Response

账户可用的模型。

Note

显示 JSON 架构：

{
  "$defs": {
    "ModelMetadata": {
      "description": "描述单个可用模型的元数据。",
      "properties": {
        "name": {
          "title": "名称",
          "type": "string"
        },
        "description": {
          "title": "描述",
          "type": "string"
        },
        "release_date": {
          "title": "发布日期",
          "type": "string"
        }
      },
      "required": [
        "name",
        "description",
        "release_date"
      ],
      "title": "ModelMetadata",
      "type": "object"
    }
  },
  "description": "账户可用的模型。",
  "properties": {
    "models": {
      "items": {
        "$ref": "#/$defs/ModelMetadata"
      },
      "title": "模型",
      "type": "array"
    }
  },
  "required": [
    "models"
  ],
  "title": "ListModelsResponse",
  "type": "object"
}

字段：

models (tuple[ModelMetadata, ...])

request_id

cached property

request_id: str

x-typesafe-request-id 响应头。

raw_http_response

property

raw_http_response: httpx2.Response

底层的 httpx2.Response，暴露状态、标头和正文。

model_config

class-attribute instance-attribute

model_config = ConfigDict(
    extra="ignore", frozen=True, strict=True
)

models

pydantic-field

models: tuple[ModelMetadata, ...]

可用的模型。

typesafe_sdk.ModelMetadata

pydantic-model

基类：Schema

描述单个可用模型的元数据。

Note

显示 JSON 架构：

{
  "description": "描述单个可用元数据的元数据。",
  "properties": {
    "name": {
      "title": "名称",
      "type": "字符串"
    },
    "description": {
      "title": "描述",
      "type": "字符串"
    },
    "release_date": {
      "title": "发布日期",
      "type": "字符串"
    }
  },
  "required": [
    "名称",
    "描述",
    "发布日期"
  ],
  "title": "ModelMetadata",
  "type": "对象"
}

字段：

name (str)
description (str)
release_date (str)

name

pydantic-field

name: str

请求的 model 字段接受的模型名称或别名。

description

pydantic-field

description: str

模型及其功能的人类可读描述。

release_date

pydantic-field

release_date: str

模型的发布日期，格式为 YYYY-MM-DD。