2.3 KiB
Mini LLM Flow - LLM Framework in 100 Lines
A 100-line minimalist LLM framework for agents, task decomposition, RAG, etc.
-
Install via
pip install minillmflow, or just copy the source (only 100 lines) -
Pro tip: Build LLM apps with LLMs assistants (ChatGPT, Claude, etc.)
-
GPT assistant: Check out Mini LLM Flow Assistant
-
Claude assistant: Create a project and dump the docs
-
Use your own LLMs and provide contexts via this prompt
-
Documentation: https://minillmflow.github.io/miniLLMFlow/
Why Mini LLM Flow?
Mini LLM Flow is designed to be the framework used by LLMs. In the future, LLM projects will self-programmed by LLMs themselves: Users specify requirements, and LLMs will design, build, and maintain. Current LLMs are:
-
👍 Good at Low-level Details: LLMs can handle LLM wrappers, tools, and prompts, which don't require any framework. Current frameworks are often over-engineered, making them difficult for humans (and LLMs) to understand.
-
👎 Bad at High-level Paradigms: While paradigms like MapReduce, task decomposition, and agents are powerful, LLMs still struggle to design them elegantly. These high-level concepts should be emphasized in frameworks.
The ideal framework for LLMs should (1) strip away low-level implementation details, and (2) keep high-level paradigms to program against. Hence, we provide this minimal (100-line) framework that allows LLMs to focus on what matters.
Mini LLM Flow is also a great learning resource, as current frameworks abstract too much away.
Tutorial
-
Documentation: https://minillmflow.github.io/miniLLMFlow/
-
Beginner Tutorial: Text summarization for Paul Graham Essay + QA agent
- Have questions for this tutorial? Ask LLM assistants through this prompt