Модели /

Microsoft: Microsoft-Decision-1

Идентификатор модели

microsoft/microsoft-decision-1

Характеристики

Контекстное окно
33K токенов
Провайдер
32768
Архитектура
text->decisions

Модальности

Входящие

  • text

Исходящие

  • decisions

Цены

Оплата по факту за использованные токены и операции. Цены в рублях (за 1 млн токенов или за единицу — как указано).

Тип Стоимость
Входящие токены 10.00 ₽ / 1M

Цены могут меняться в зависимости от курса валют. Данные по тарифам поступают от поставщика услуг и обновляются каждые сутки.

Начать использовать

Описание

Here's a thinking process:

  1. Analyze User Request:
  • Task: Translate a model description from English to Russian.
  • Source: A catalog description of an AI model.
  • Constraints:
    • Keep model names and technical terms in the familiar format for Russian-speaking developers (this usually means keeping English names/terms as-is, or transliterating if standard, but typically keeping them in English in tech contexts).
    • Output only the translated text, no introduction, no quotes around the whole text.
  • Input:
    • Model line: "Модель: Microsoft: Microsoft-Decision-1 (microsoft/microsoft-decision-1)"
    • English text: "Microsoft-Decision-1 is a small model built for fast decision-making. Instead of generating text, it reads the provided content and returns a calibrated probability for each fixed answer option, so the..."
  1. Identify Key Parts:
  • Model name: Microsoft-Decision-1 / microsoft/microsoft-decision-1 -> Keep as is (standard practice in Russian dev contexts to keep English identifiers).
  • The description text starts with: "Microsoft-Decision-1 is a small model built for fast decision-making. Instead of generating text, it reads the provided content and returns a calibrated probability for each fixed answer option, so the..."
  1. Translate the Text (while respecting constraints):
  • Need to translate the meaning into Russian, but keep the model name and technical terms in English (as they are "familiar for Russian-speaking developers").
  • Phrases to translate:
    • "is a small model built for fast decision-making" -> "маленькая модель, предназначенная для быстрого принятия решений"
    • "Instead of generating text" -> "Вместо генерации текста"
    • "it reads the provided content" -> "она читает предоставленное содержимое" or "обрабатывает предоставленный контент"
    • "returns a calibrated probability for each fixed answer option" -> "возвращает калиброванную вероятность для каждого фиксированного варианта ответа"
    • "so the..." -> The text ends with "so the...". I need to handle the incomplete sentence. Since the user provided it ending with "so the...", I should translate what's there, and if it's incomplete, I might need to either complete it reasonably or just translate up to the end of the provided text. Looking at the input: "so the..." - it's cut off. I should translate the full visible part, and if needed, indicate it's incomplete, but the instruction says "Translate the description model from catalog." I'll translate the given text, keeping the sentence structure logical in Russian, and since it ends with "so the...", I'll translate up to that point, maybe keeping the ellipsis or completing it naturally if possible, but better to stick to the text provided. Actually, the user gave: "so the..." as the very end. I'll translate the whole block, and since it ends abruptly, I'll either keep the ellipsis in Russian or just translate what's there. Let me read carefully: "Microsoft-Decision-1 is a small model built for fast decision-making. Instead of generating text, it reads the provided content and returns a calibrated probability for each fixed answer option, so the..." -> I'll translate it fully, and since it ends with "so the...", I'll output the translation ending similarly or completing it if context allows, but to be safe, I'll translate the exact text provided, keeping the ellipsis or indicating it's cut off. However, the instruction says "Ответь только текстом перевода, без вступления и без кавычек вокруг всего текста." I'll just provide the Russian translation of the visible text, keeping the model name in English as requested.

Let me draft the translation: "Microsoft-Decision-1 is a small model built for fast decision-making. Вместо генерации текста она читает предоставленное содержимое и возвращает калиброванную вероятность для каждого фиксированного варианта ответа, так что..."

But wait, the user said: "Saves model names and technical terms in the familiar for Russian-speaking developers view." This means I should keep Microsoft-Decision-1 and microsoft/microsoft-decision-1 as is, which I'm already doing by not translating them.

Also, the output should be "only the text translation, no introduction, no quotes around the whole text."

Let me produce the exact output format expected. The input has: "Модель: Microsoft: Microsoft-Decision-1 (microsoft/microsoft-decision-1)" Then the English text.

I need to output the translated description. The user probably wants the whole thing in Russian, with the model line kept as is or also translated? The instruction: "Saves model names and technical terms in the familiar for Russian-speaking developers view." This likely means keep them in English. I'll keep the model line exactly as given, or maybe just the description part. Let me see the structure. The user provided: "Модель: Microsoft: Microsoft-Decision-1 (microsoft/microsof…

Решения

Модель не пишет текст. В state передаются данные, в questions — вопросы с известными вариантами. Ответ приходит готовыми значениями.

  • noul — да или нет, вероятность от 0 до 1
  • choice — один вариант из объекта criteria
  • score — положение на шкале, уровни задаются массивом criteria
import requests

response = requests.post(
    "https://routerapi.ru/api/v1/decisions",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "microsoft/microsoft-decision-1",
        "state": "С 10:40 все запросы к API возвращают 500.",
        "questions": {
            "is_urgent": {
                "type": "noul",
                "instructions": "Требует ли обращение немедленной реакции?",
                "criteria": {"true": "Сервис недоступен", "false": "Можно разобрать в очереди"},
            },
            "team": {
                "type": "choice",
                "instructions": "На какую команду направить обращение?",
                "criteria": {"billing": "Оплата и счета", "technical": "Ошибки API", "other": "Всё остальное"},
            },
            "severity": {
                "type": "score",
                "instructions": "Насколько сильно это влияет на работу клиента?",
                "criteria": ["Влияния нет", "Есть обходной путь", "Работа заблокирована"],
            },
        },
    },
)
answers = response.json()["answers"]
print(answers["is_urgent"]["noul"], answers["team"]["choice"], answers["severity"]["score"])

Ответы приходят в answers под теми же ключами, что и вопросы.