NVIDIA-NeMo/labs-OO-Agents

Health v2 · maintenance only

Flags: young no_license

Adoption not part of the score

1906 stars · 263 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NVIDIA Object Oriented Agents (NOOA) is a model-agnostic Python framework for building AI agents where state, tools, prompts, and typed interfaces are expressed through a single Python class. It provides an object-oriented alternative to fragmented agent abstractions, letting developers define agentic methods that the runtime hands to an LLM.

Use cases

• build ai agents in python
• create a support agent that triages tickets with an llm
• define agent tools as typed python methods
• build llm-powered workflows with typed state
• prototype agentic applications without learning a dsl
• swap underlying llm models without rewriting agent code

When to choose

• you want agents expressed as plain python classes with typed fields and methods
• you prefer object-oriented design over prompt/callback-based agent frameworks
• you need model-agnostic agent development
• you want strong typing on agent inputs and outputs

When to avoid

• you need a visual or low-code agent builder
• you require a framework with a large plugin ecosystem and long track record
• you are building simple single-prompt llm calls that don't need an agent framework

Facets

framework · maturity active

agent-framework llm-inference sdk artificial-intelligence large-language-models developer-tools python cross-platform object-oriented-agents nvidia typed-agents model-agnostic llm ai-agents

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVIDIA-NeMo/labs-OO-Agents")