EdgeChains  by arakoodev

Full-stack GenAI library for declarative prompt/chain management

created 2 years ago
396 stars

Top 74.0% on sourcepulse

GitHubView on GitHub
1 Expert Loves This Project
Project Summary

EdgeChains.js is a full-stack Generative AI library designed for managing complex prompt engineering, distributed computing, and API integrations. It targets developers seeking a declarative, configuration-driven approach to building GenAI applications, offering a unified solution for front-end, back-end, and prompt management.

How It Works

EdgeChains leverages jsonnet for declarative configuration of prompts and chains, treating GenAI as a deployment and configuration challenge akin to Kubernetes. This approach, inspired by Google's Borg infrastructure management, allows for versionable, diffable prompts and automatic parallelism across CPUs, GPUs, and TPUs via a WebAssembly runtime. It emphasizes fault tolerance and scalability for complex chain-of-thought applications.

Quick Start & Requirements

  • Install dependencies with npm install within the example directory (e.g., JS/edgechains/examples/chat-with-pdf/).
  • Requires Node.js, npm, and a PostgreSQL Vector database (Supabase is demonstrated).
  • API keys for services like OpenAI and Supabase are needed.
  • Setup involves cloning the repository, installing packages, configuring secrets, and setting up the database schema.
  • Example: ChatWithPdfExample

Highlighted Details

  • Declarative prompt and chain management using jsonnet.
  • Automatic parallelism and fault tolerance for LLM chains.
  • Built on Hono.js and jsonnet, citing stability from Cloudflare and Google.
  • Integrated token cost measurement and tracking.

Maintenance & Community

  • Active community with a Discord server mentioned for support.
  • Acknowledges inspiration from Next.js and contributions from the LLM community.
  • Contribution guidelines and a code of conduct are provided.

Licensing & Compatibility

  • Licensed under GNU Affero General Public License v3.0.
  • Commercial licensing is available upon contact.

Limitations & Caveats

The project is licensed under AGPLv3, which may impose copyleft restrictions on derivative works, potentially impacting commercial use or integration into closed-source projects without specific licensing agreements.

Health Check
Last commit

2 days ago

Responsiveness

1 day

Pull Requests (30d)
8
Issues (30d)
0
Star History
12 stars in the last 90 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of AI Engineering, Designing Machine Learning Systems) and Travis Fischer Travis Fischer(Founder of Agentic).

pezzo by pezzolabs

0.2%
3k
Open-source LLMOps platform for streamlining AI workflows
created 2 years ago
updated 1 month ago
Feedback? Help us improve.