Skip to content
@Pipelex

Pipelex

The language for declarative composable Al workflows. Devtool for agents and mere humans.
Pipelex Logo

Executable AI Methods

Declare multi-step AI methods in typed .mthds files — deterministic orchestration, structured outputs, repeatable results.


MIT License PyPI Discord


What a Method Looks Like

[pipe.summarize_article]
type    = "PipeLLM"
inputs  = { article = "Text", audience = "Text" }
output  = "Text"
prompt  = "Summarize $article in three bullet points for $audience."

From here, Pipelex handles model routing across 60+ models, structured output parsing, and pipeline orchestration.


Get Started with Claude Code

Install the mthds CLI, it's required by the plugin for method validation and execution:

npm install -g mthds

Start Claude Code:

claude

Tell Claude to install the MTHDS plugins marketplace:

/plugin marketplace add mthds-ai/mthds-plugins

Then install the MTHDS Claude Code plugin:

/plugin install mthds@mthds-plugins

Then reload plugins:

/reload-plugins

If that doesn't work, exit Claude Code and reopen it:

/exit

Build your first method:

/mthds-build A method to summarize articles with key takeaways for different audiences

Run it:

/mthds-run

Get Started without Claude Code

uv tool install pipelex
pipelex init
  1. Install the VS Code extension for .mthds syntax highlighting
  2. Browse methods on the MTHDS Hub for inspiration
  3. Author your own .mthds methods
  4. Validate with pipelex validate bundle your_method.mthds
  5. Run with pipelex run bundle your_method.mthds

Configure AI Access

  • Pipelex Gateway (Recommended) — Free credits, single API key for LLMs, OCR / document extraction, and image generation across all major providers.
  • Bring Your Own Keys — Use existing API keys from OpenAI, Anthropic, Google, Mistral, etc. See Configure AI Providers.
  • Local AI — Ollama, vLLM, LM Studio, or llama.cpp — no API keys required. See Configure AI Providers.

Examples

Ready-to-run methods in the Cookbook: classification, extraction, analysis, generation, and more.

Community & Support

Key Repositories

Repository Description
pipelex Python runtime — build and run AI methods
mthds The MTHDS open standard — specification and docs
mthds-plugins Claude Code plugin — skills for building, running, and editing methods
pipelex-cookbook Production-ready examples and tutorials

License

All repositories are MIT licensed unless otherwise specified. See individual LICENSE files for details.

"Pipelex" is a trademark of Evotis S.A.S.

Pinned Loading

  1. pipelex pipelex Public

    Declarative language for composable Al workflows. Devtool for agents and mere humans.

    Python 639 45

  2. pipelex-cookbook pipelex-cookbook Public

    Cookbook for Pipelex, the declarative language for composable Al workflows. Devtool for agents and mere humans.

    Makefile 35 9

  3. pipelex-mcp pipelex-mcp Public

    MCP server to run Pipelex pipelines

    Makefile 8 1

  4. pipelex-starter pipelex-starter Public template

    Simple starter template to get you up and running with a project to build and run Pipelex pipelines

    Makefile 5 1

  5. pipelex-api pipelex-api Public

    API of Pipelex: open-source language for AI Agents to create and run repeatable AI workflows

    Python 2

  6. n8n-nodes-pipelex n8n-nodes-pipelex Public

    n8n Node for running Pipelex

    Makefile

Repositories

Showing 10 of 16 repositories

Top languages

Loading…

Most used topics

Loading…