YEScale Chat Model icon

YEScale Chat Model

YEScale chat model for AI Agent, supporting 4 endpoint modes

YEScale Chat Model

AI-generated

Summary

Provide a configured YEScale Chat Model instance for AI interactions using one of four supported API modes (Chat Completions, Claude Messages, Gemini Native, or Responses API).

Inputs

  • API Mode (required) — Select the API mode to use for chat model interactions: 'chatCompletions', 'claudeMessages', 'geminiNative', or 'responses'.
  • Model (required) — Choose the specific model compatible with the selected API mode; presented options dynamically load based on the API mode.
  • Other Params — Optional parameters such as Max Tokens (minimum 1) and Temperature (range 0 to 2) to influence response generation.
  • Timeout — Node-level request timeout in milliseconds (minimum 1000; default 200000).
  • Max Retries — Node-level number of retry attempts on failure (minimum 0; default 3).
  • Built-In Tools — For 'responses', 'claudeMessages', and 'geminiNative' modes, select supported built-in tools such as Web Search or Web Fetch.
  • Reasoning — For supported modes, optional reasoning configuration including Support Effort level, loaded from the model's capabilities.

Output shape

a single AI language model instance configured for the selected YEScale chat API mode and model

The node outputs a configured chat model instance for downstream nodes to consume, corresponding to the selected API mode and model. It dynamically loads compatible models and tools based on the API mode. The model supports customizable parameters like temperature and max tokens, and includes built-in support for reasoning and external tools depending on mode and model capabilities.

Links

Discussion