VolcengineAi Chat Model icon

VolcengineAi Chat Model

For advanced usage with an AI chain

VolcengineAi Chat Model

AI-generated

Summary

Select a Doubao model and configure API options, then invoke the model to generate a response.

Inputs

  • model (required) — Select the Doubao model to use, either from the list or by ID.
  • options.baseURL — Override the default API base URL for the request.
  • options.responseFormat — Response format; set to json_object for JSON mode.
  • options.temperature — Controls randomness of the output.
  • options.maxTokens — Maximum number of tokens to generate.
  • options.topP — Controls diversity via nucleus sampling.
  • options.frequencyPenalty — Penalizes token repetition.
  • options.presencePenalty — Penalizes repeated topics.
  • options.reasoningEffort — Level of reasoning effort to use.
  • options.maxRetries — Maximum number of retries on failure.

Output shape

a single record

The response contains the model's generated text or JSON; may include error details.

Examples

Example 1: Default model with text response

model=doubao-pro-4k, responseFormat=text

Example 2: JSON mode with low temperature

model=doubao-chat, responseFormat=json_object, temperature=0.2

Example 3: Base URL override

baseURL=https://custom.volces.com/api/v3, model=doubao-code

Discussion