Skip to content

ParaLLeM

ParaLLeM is a general-purpose, batch-first Python library for building LLM workflows in ordinary Python control flow.

Why use it?

  • Switch between sync, concurrent, and batch execution with just 1 line of code.
  • Write applications in Python instead of a data structure or domain-specific language.
  • Supports agentic features like tool calls, structured output, image input, and web search.
  • Saves responses to disk for fast iteration.

Installation

pip install parallem

Small example

from dotenv import load_dotenv
import parallem as pllm

load_dotenv()

with pllm.resume_directory(
    ".pllm/simplest",
    provider="openai",
    strategy="sync",
    dashboard=True,
) as orch:
    with orch.agent() as agt:
        resp = agt.ask_llm("Please name a power of 3.", hash_by=["llm"])
        agt.print(resp.final_answer)

[INFO] Resuming with session_id=0
27 (which is 3^3).
[DASH] ↘ b14ccd95
  • Start with the quickstart for the main workflow.
  • Compare sync, concurrent, and batch in the strategy guide.
  • Read the philosophy page for the design goals behind the library.