{"id":24133,"library":"nvidia-nat","title":"NVIDIA NeMo Agent Toolkit (nvidia-nat)","description":"NVIDIA NeMo Agent Toolkit (nvidia-nat) is a Python library for building, training, and deploying conversational AI agents, including large language model (LLM) agents with tool use, memory, and multi-turn dialogue. Version 1.6.0 requires Python <3.14, >=3.11. Releases occur approximately monthly.","status":"active","version":"1.6.0","language":"python","source_language":"en","source_url":"https://github.com/NVIDIA/NeMo-Agent-Toolkit","tags":["NVIDIA","NeMo","agent","LLM","conversational AI","toolkit"],"install":[{"cmd":"pip install nvidia-nat","lang":"bash","label":"Install from PyPI"},{"cmd":"pip install nvidia-nat[all]","lang":"bash","label":"Install with all extras (e.g., local models, streaming)"}],"dependencies":[{"reason":"Required for model inference and training; must be installed separately with CUDA support","package":"torch","optional":false},{"reason":"Used for Hugging Face model integration; optional if using only local Nemo models","package":"transformers","optional":true}],"imports":[{"note":"Primary class for creating an agent.","symbol":"NemoAgent","correct":"from nvidia_nat import NemoAgent"},{"note":"Configuration class for agent settings.","symbol":"AgentConfig","correct":"from nvidia_nat.config import AgentConfig"}],"quickstart":{"code":"import os\nfrom nvidia_nat import NemoAgent\nfrom nvidia_nat.config import AgentConfig\n\napi_key = os.environ.get('NEMO_API_KEY', '')\nconfig = AgentConfig(api_key=api_key)\nagent = NemoAgent(config)\nresponse = agent.chat(\"Hello, how are you?\")\nprint(response)","lang":"python","description":"Initialize a NemoAgent with an API key from environment variable and send a chat message."},"warnings":[{"fix":"Update imports to 'from nvidia_nat.config import AgentConfig'.","message":"In version 1.6.0, the AgentConfig class is no longer defined in nvidia_nat directly; it must be imported from nvidia_nat.config. Old imports (from nvidia_nat import AgentConfig) raise ImportError.","severity":"breaking","affected_versions":"<=1.5.x"},{"fix":"Replace agent.call() with agent.chat().","message":"The method 'agent.call()' is deprecated in 1.6.0 and will be removed in 2.0. Use 'agent.chat()' instead.","severity":"deprecated","affected_versions":">=1.6.0"},{"fix":"Install PyTorch with CUDA: pip install torch --index-url https://download.pytorch.org/whl/cu118","message":"CUDA is required for GPU acceleration. If torch is installed without CUDA, the library silently falls back to CPU, which is extremely slow for large models.","severity":"gotcha","affected_versions":"all"}],"env_vars":null,"last_verified":"2026-05-01T00:00:00.000Z","next_check":"2026-07-30T00:00:00.000Z","problems":[{"fix":"Use: from nvidia_nat.config import AgentConfig","cause":"AgentConfig was moved to nvidia_nat.config in version 1.6.0.","error":"ImportError: cannot import name 'AgentConfig' from 'nvidia_nat'"},{"fix":"Run: pip install nvidia-nat","cause":"Library not installed or installed in wrong environment.","error":"ModuleNotFoundError: No module named 'nvidia_nat'"},{"fix":"Set the environment variable, e.g., export NEMO_API_KEY='your_key' or pass api_key parameter to AgentConfig.","cause":"The library requires an API key for authentication.","error":"RuntimeError: NEMO_API_KEY environment variable not set"}],"ecosystem":"pypi","meta_description":null,"install_score":null,"install_tag":null,"quickstart_score":null,"quickstart_tag":null}