https://github.com/kdeps/kdeps

Kdeps reduces the complexity of building self-hosted RAG AI Agents and APIs powered by open-source LLMs
https://github.com/kdeps/kdeps

agents ai-agents api artificial-intelligence cuda docker dockerized fine-tuning huggingface llama llm llm-agent llmops mistral mlops model-inference multimodal nvidia opensource vicuna

Last synced: 2 days ago
JSON representation

Kdeps reduces the complexity of building self-hosted RAG AI Agents and APIs powered by open-source LLMs

• Host: GitHub
• URL: https://github.com/kdeps/kdeps
• Owner: kdeps
• License: apache-2.0
• Created: 2024-08-01T01:46:41.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2025-01-28T08:21:29.000Z (7 days ago)
• Last Synced: 2025-01-28T09:24:59.978Z (7 days ago)
• Topics: agents, ai-agents, api, artificial-intelligence, cuda, docker, dockerized, fine-tuning, huggingface, llama, llm, llm-agent, llmops, mistral, mlops, model-inference, multimodal, nvidia, opensource, vicuna
• Language: Go
• Homepage: https://kdeps.com
• Size: 3.72 MB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Security: SECURITY.md

Awesome Lists containing this project

README

        
Kdeps is a multi-model AI agent framework that is optimized for creating purpose-built

Dockerized AI agent APIs ready to be deployed on/off cloud. It utilizes self-contained

open-source LLM models that are orchestrated by a graph-based dependency workflow.



Kdeps is packed with features:

  - 🚀 Run Kdeps in single-execution or API Mode

  - 🤖 Use Multiple Open-Source LLMs

  - 🐍 Run Python scripts in isolated environments using Anaconda

  - 🖥️ Execute Custom Shell-Scripts

  - 🧪 Anaconda Support

  - 🔄 Share and Remix AI Agents

  - 🌐 Interact with external HTTP APIs directly into the resource

  - 📊 Generate structured outputs from LLMs

  - 📦 Install dependent Ubuntu packages from within the workflow configuration

  - 📜 Define custom Ubuntu repositories and PPAs in the workflow

  - 📈 RAG Graph-based workflow execution

  - ⚡ Written in Golang

## Installation

Please see the installation instructions [here](https://kdeps.github.io/kdeps/getting-started/introduction/installation.html).

## Quickstart

Take a look at our quickstart guide [here](https://kdeps.github.io/kdeps/getting-started/introduction/quickstart.html).