Kenhuang Academy
  • English

      • Installation

      Requirements

      • Python 3.10+ (3.11 / 3.12 recommended)
      • pip, uv, or poetry
      • At least one LLM provider API key

      Virtual environment

      python -m venv .venv
# Windows
.venv\Scripts\activate
# macOS / Linux
source .venv/bin/activate

      Install packages

      LangChain 1.0 splits core and integrations. A minimal agent setup:

      pip install -U langchain langgraph langchain-openai

      Other providers:

      pip install langchain-anthropic
pip install langchain-google-genai
pip install langchain-community   # loaders, some vector stores

      Verify:

      python -c "import langchain; import langgraph; print('ok')"

      API keys

      Environment variables (recommended)

      export OPENAI_API_KEY="sk-..."
export ANTHROPIC_API_KEY="sk-ant-..."

      .env file

      pip install python-dotenv
      OPENAI_API_KEY=sk-...
      from dotenv import load_dotenv
load_dotenv()

      Never commit .env to Git.

      Model identifiers

      Use provider:model strings:

      "openai:gpt-4.1"
"anthropic:claude-sonnet-4-6"
"google_genai:gemini-2.0-flash"

      See Models & Messages for initialized model instances.

      LangSmith (optional)

      For tracing and evals:

      export LANGCHAIN_TRACING_V2=true
export LANGCHAIN_API_KEY="lsv2_..."
export LANGCHAIN_PROJECT="my-project"

      See Production & LangSmith.

      Suggested layout

      my-langchain-app/
├── .env
├── src/agents/
├── src/tools/
└── tests/

      Troubleshooting

      ModuleNotFoundError: langchain.agents?
      Install langchain ≥ 1.0.

      Connection errors?
      Check network/proxy; try another provider or local Ollama (langchain-ollama).

      Dependency conflicts?
      Use a fresh venv.

      Next steps