Works with LangChain

LangChain, running on your Mac

ToolPiper is a LangChain chat model. Everything you already build with `ChatOpenAI` works against a model running on your own Mac.

Download ToolPiper Read the API docs

Free, Apple Silicon, signed. The local model runner has no caps and needs no account.

Setup

  1. Install ToolPiper and launch it once. The local API listens on port 9998.
  2. Install the package: `pip install langchain-toolpiper`.
  3. Point `ChatToolPiper` at a model that is resident locally.
from langchain_toolpiper import ChatToolPiper

# No API key, no base URL: the package discovers the local server.
llm = ChatToolPiper(model="gemma-4-12b-it-Q4_K_M")
print(llm.invoke("Summarise this in one sentence.").content)

Prompts, responses and files stay on the machine. There is no key to rotate and no per-token cost. LangChain talks to ToolPiper over its local OpenAI-compatible API on port 9998.