Installation

Install Mellea and set up your Python environment.

Prerequisites: Python 3.11+, pip or uv available.

Install

pip install mellea

uv add mellea

Note: uv add adds Mellea to an existing uv project, so run it from a directory created with uv init (or one that already has a pyproject.toml). To install into a plain virtual environment without a uv project, use uv pip install mellea instead.

Optional extras

Install extras for specific backends and features:

pip install "mellea[litellm]"    # LiteLLM multi-provider (Anthropic, Bedrock, etc.)
pip install "mellea[hf]"         # HuggingFace transformers for local inference
pip install "mellea[watsonx]"    # IBM WatsonX
pip install "mellea[tools]"      # Tool and agent dependencies (LangChain, smolagents)
pip install "mellea[cli]"        # m serve, m alora, m decompose CLI commands
pip install "mellea[telemetry]"  # OpenTelemetry tracing and metrics

uv add "mellea[litellm]"        # LiteLLM multi-provider (Anthropic, Bedrock, etc.)
uv add "mellea[hf]"             # HuggingFace transformers for local inference
uv add "mellea[watsonx]"        # IBM WatsonX
uv add "mellea[tools]"          # Tool and agent dependencies (LangChain, smolagents)
uv add "mellea[cli]"            # m serve, m alora, m decompose CLI commands
uv add "mellea[telemetry]"      # OpenTelemetry tracing and metrics

You can combine extras:

pip install "mellea[litellm,tools,telemetry]"

uv add "mellea[litellm,tools,telemetry]"

All extras: mellea[all] installs everything. For the full list of available extras see pyproject.toml.

Default backend: Ollama

The default session connects to Ollama running locally. Install Ollama and pull the default model before running any examples:

ollama pull granite4.1:3b