Skip to main content

gpt_4o

What it is​

  • A module-level definition of a ChatModel configured for OpenAI’s gpt-4o via langchain_openai.ChatOpenAI.
  • Intended to be imported and used as a ready-to-use chat model instance.

Public API​

  • ID: str
    • Model identifier: "gpt-4o".
  • PROVIDER: str
    • Provider identifier: "openai".
  • model: naas_abi_core.models.Model.ChatModel
    • Preconfigured ChatModel wrapping a ChatOpenAI instance:
      • model="gpt-4o"
      • temperature=0
      • api_key sourced from ABIModule.get_instance().configuration.openai_api_key (wrapped as pydantic.SecretStr)

Configuration/Dependencies​

  • Depends on:
    • langchain_openai.ChatOpenAI
    • naas_abi_core.models.Model.ChatModel
    • naas_abi_marketplace.ai.chatgpt.ABIModule for configuration access
    • pydantic.SecretStr
  • Required configuration:
    • ABIModule.get_instance().configuration.openai_api_key must be set to a valid OpenAI API key.

Usage​

from naas_abi_marketplace.ai.chatgpt.models.gpt_4o import model

# `model` is a ChatModel that wraps a LangChain ChatOpenAI client.
# How you call it depends on the ChatModel interface in naas_abi_core.
print(model.model_id, model.provider)

Caveats​

  • Importing this module will instantiate ChatOpenAI immediately and read the OpenAI API key from ABIModule configuration. If configuration is missing or not initialized, import-time errors may occur.