Skip to main content

gemma3_4b

What it is​

  • A module that defines a preconfigured ChatModel instance for the Gemma3 4B model served via Ollama (LangChain ChatOllama backend).

Public API​

  • model: ChatModel
    • Ready-to-use chat model configuration:
      • model_id: "gemma3:4b"
      • name: "Gemma3 4B"
      • description: "Google's open-source Gemma3 4B model for local deployment. Fast, lightweight alternative to cloud Gemini."
      • provider: "ollama"
      • image: Ollama logo URL
      • context_window: 8192
      • Underlying LangChain model: ChatOllama(model="gemma3:4b", temperature=0.4)

Configuration/Dependencies​

  • Dependencies:
    • langchain_ollama.ChatOllama
    • naas_abi_core.models.Model.ChatModel
  • Runtime requirement:
    • An Ollama setup capable of serving the model id "gemma3:4b".

Usage​

from naas_abi_marketplace.ai.gemma.models.gemma3_4b import model

# Access underlying LangChain chat model
llm = model.model

# Example call (LangChain interface)
response = llm.invoke("Hello! Give me a one-sentence summary of Gemma.")
print(response)

Caveats​

  • The module only exports a configured ChatModel instance; it does not include Ollama installation, model pulling, or server lifecycle management.