Skip to main content

Getting Started

Prerequisites​

  • Python 3.13 (Poetry recommended)
  • Optional: Docker, Ollama or LM Studio

Set up a local LLM provider (recommended): Local LLM providers

Install​

# Add core library to your project
poetry add l6e-forge

# Add CLI as a dev dependency
poetry add --group dev l6e-forge-cli

Using uv:

uv add l6e-forge
uv add --dev l6e-forge-cli

Using pip:

pip install l6e-forge l6e-forge-cli

Create a Workspace​

poetry run forge init ./my-workspace
cd my-workspace

Create an Agent​

poetry run forge create agent my-assistant --template=assistant
poetry run forge create agent my-ollama --provider=ollama --model llama3.2:3b

Bootstrap Models​

poetry run forge models bootstrap agents/my-ollama --provider-order ollama,lmstudio --interactive

See more details on recommendations, flags, and how it updates your agent: Auto model selection & bootstrapping

Run the Stack​

poetry run forge up

API: http://localhost:8000 — Monitor: http://localhost:8321 — UI: http://localhost:8173

Chat​

poetry run forge chat my-ollama -w ./my-workspace

Workspace Structure​

my-workspace/
├── forge.toml
├── agents/
│   └── my-ollama/
│       ├── agent.py
│       ├── config.toml
│       └── tools.py (optional)
└── .forge/
    ├── logs/
    └── data/

Next steps​