rephrase readme

This commit is contained in:
zachary62 2024-12-26 21:59:18 +00:00
parent cab0ab6a1c
commit 87929060c8
2 changed files with 15 additions and 12 deletions

3
.gitignore vendored
View File

@ -69,3 +69,6 @@ htmlcov/
*.bak *.bak
*.tmp *.tmp
*.temp *.temp
test.ipynb

View File

@ -5,24 +5,24 @@
Minimalist LLM framework in [100 lines](minillmflow/__init__.py). Express popular paradigms like agents, task decomposition, RAG, and more. Minimalist LLM framework in [100 lines](minillmflow/__init__.py). Express popular paradigms like agents, task decomposition, RAG, and more.
- Install via ```pip install minillmflow```. Or just copy the [source codes](minillmflow/__init__.py) (it's only 100 lines) - Install via ```pip install minillmflow```. Or just copy the [source codes](minillmflow/__init__.py) (it's only 100 lines)
- We **strongly recommened** using LLMs (e.g., Claude, ChatGPT, Cursor) to build LLM applications with [this prompt](minillmflow/docs/prompt) - We **strongly recommened** using LLMs (e.g., Claude, ChatGPT, Cursor) to develop LLM applications with [this prompt](minillmflow/docs/prompt)
## Why miniLLMFlow? ## Why miniLLMFlow?
The future of programming will be heavily LLM-assited, and LLMs: The future of programming will be heavily LLM-assited, and LLMs:
1. **👍 Excel at Feature Implementation**: 1. **😀 Shine at Feature Implementation**:
Give proper context/docs, they know how to use APIs, tools, text chunking, prompt wrapping, etc. With proper docs, LLMs can handle APIs, tools, chunking, prompt wrapping, etc.
These shouldn't be part of an LLM framework, as they're hard to maintain, update, and optimize. These are hard to maintain and optimize, so they dont belong in the framework.
2. **👎 Suck at Paradigm Design**: 2. **☹️ Struggle with Paradigm Design**:
Paradigms like MapReduce, DAG workflows, and recent agents are powerful for reasoning about problems. Paradigms like MapReduce, task decomposition, and agents are powerful, even for LLMs.
However, designing elegant paradigms is challenging, and LLMs tend to write redundant code. However, designing these elegantly remains challenging for LLMs.
Can we build a framework that: To enable LLMs to develop LLM applications, a framework should
(1) removes redunant feature implementations, (1) remove redunant feature implementations but
(2) but keep core paradigms for LLMs to program against? (2) keep core paradigms to build on.
Turns out that we just need 100 lines of codes. It turns out that such a framework only needs 100 lines of code.
<div align="center"> <div align="center">
<img src="./docs/minillmflow.jpg" width="400"/> <img src="./docs/minillmflow.jpg" width="400"/>