Home
Last updated
Last updated
A minimalist AI framework for Agents, Task Decomposition, RAG, etc.
Lightweight: Just the core graph abstraction in 65 lines. ZERO dependencies, and vendor lock-in.
Expressive: Everything you love from larger frameworks—(), , , and more.
Agentic-Coding: Intuitive enough for AI agents to help humans build complex LLM applications.
We model the LLM workflow as a Graph + Shared Store:
From there, it’s easy to implement popular design patterns:
We do not provide built-in utilities. Instead, we offer examples—please implement your own:
Why not built-in?: I believe it's a bad practice for vendor-specific APIs in a general framework:
API Volatility: Frequent changes lead to heavy maintenance for hardcoded APIs.
Flexibility: You may want to switch vendors, use fine-tuned models, or run them locally.
Optimizations: Prompt caching, batching, and streaming are easier without vendor lock-in.
handles simple (LLM) tasks.
connects nodes through Actions (labeled edges).
enables communication between nodes within flows.
nodes/flows allow for data-intensive tasks.
helps manage concurrency and rate limits.
autonomously makes decisions.
chains multiple tasks into pipelines.
integrates data retrieval with generation.
splits data tasks into Map and Reduce steps.
formats outputs consistently.
coordinate multiple agents.
Check out , the fastest way to develop LLM projects with Brainy Flow!
