Skip to main content

airgap_qwen

What it is​

  • A module that defines a preconfigured ChatModel instance for the Qwen3 model running in “airgap” mode via a local Docker Model Runner endpoint.

Public API​

  • Module constants
    • MODEL_ID: "ai/qwen3"
    • PROVIDER: "qwen"
    • NAME: "qwen3-airgap"
    • DESCRIPTION: Human-readable description of the model configuration.
    • IMAGE: URL for a representative image.
    • CONTEXT_WINDOW: 8192 (declared, not wired into the ChatModel in this file)
  • Module variable
    • model: ChatModel: A ChatModel wrapping a langchain_openai.ChatOpenAI client configured to talk to a local endpoint.

Configuration/Dependencies​

  • Python dependencies
    • langchain_openai.ChatOpenAI
    • naas_abi_core.models.Model.ChatModel
  • Runtime dependency
    • A service reachable at: http://localhost:12434/engines/v1
  • Model configuration
    • ChatOpenAI(model="ai/qwen3", temperature=0.7, base_url="http://localhost:12434/engines/v1")

Usage​

from naas_abi.models.airgap_qwen import model

# Access metadata
print(model.model_id)     # "ai/qwen3"
print(model.provider)     # "qwen"
print(model.name)         # "qwen3-airgap"

# Access the underlying LangChain chat model/client
llm = model.model
print(llm)

Caveats​

  • The module only defines configuration; successful usage requires a compatible local server running at http://localhost:12434/engines/v1.
  • CONTEXT_WINDOW is declared but not applied to the ChatOpenAI/ChatModel configuration in this file.