AACFlow

YandexGPT

Text mit YandexGPT generieren und verarbeiten — Yandex's großes Sprachmodell

ya

YandexGPT is a large language model developed by Yandex, designed for a wide range of natural language processing tasks including text generation, summarization, classification, and question answering. It is optimized for the Russian language and available through Yandex Cloud infrastructure.

With the YandexGPT integration in AACFlow, you can:

  • Generate text: Create coherent, context-aware text responses for chatbots, content, and automation
  • Summarize content: Condense long articles, documents, or conversations into concise summaries
  • Classify text: Categorize user input, documents, or messages by topic, sentiment, or intent
  • Answer questions: Build knowledge-base Q&A systems with retrieval-augmented capabilities
  • Translate text: Leverage multilingual capabilities for cross-language communication
  • Process structured data: Extract entities, generate structured outputs, and parse documents

This integration brings YandexGPT's Russian-optimized NLP capabilities into your AACFlow automation workflows.

Nutzungsanleitung

Integrate YandexGPT into the workflow. Can generate text, summarize, classify, and answer questions. Requires Yandex Cloud API key and folder ID.

Tools

yandexgpt_complete

Generate text completion using YandexGPT

Eingabe

ParameterTypErforderlichBeschreibung
modelUristringJaYandexGPT model URI
messagesarrayJaArray of message objects (role + text)
temperaturenumberNeinGeneration temperature (0-1)
maxTokensnumberNeinMaximum tokens to generate
apiKeystringJaYandex Cloud API key
folderIdstringJaYandex Cloud folder ID

Ausgabe

ParameterTypBeschreibung
resultobjectGenerated completion object
alternativesarrayAlternative completions

yandexgpt_embed

Generate text embeddings

Eingabe

ParameterTypErforderlichBeschreibung
textstringJaText to embed
modelUristringJaEmbedding model URI
apiKeystringJaYandex Cloud API key
folderIdstringJaYandex Cloud folder ID

Ausgabe

ParameterTypBeschreibung
embeddingarrayVector embedding array
dimensionnumberEmbedding dimension

On this page

Heute mit dem Aufbau beginnen
Über 100 000 Entwickler vertrauen uns.
Die SaaS-Plattform zum Aufbau von KI-Agenten und für Ihre agentische Belegschaft.
Loslegen