Infrastructure
Your Full Stack GenAI Infrastructure. Deploy, Finetune, Manage Prompts and Generate Dataβall in one place.
SimpliML is an open-source project designed to streamline the process of building, training, and deploying machine learning models. Our mission is to lower the barriers to entry in the field of Generative AI (GenAI) and empower users with intuitive tools and comprehensive documentation.
SimpliML offers a user-friendly interface and extensive resources, making it an excellent choice for both beginners and seasoned professionals.
NOTE: SimpliML requires Kubernetes for deployment. Kubernetes provides the necessary orchestration and scaling capabilities to manage the infrastructure efficiently. Ensure you have a Kubernetes cluster set up before deploying SimpliML.
Database Deployment
The easy way to set up elasticsearch and kibana is through Helm. Follow this documentation
helm repo add elastic https://helm.elastic.co
helm repo update
SimpliML Application
The easy way to set up simpliml is through Helm. Follow this documentation
helm repo add simpliml https://quarkal-ai.github.io/simpliml-helm
helm repo update
To get started with SimpliML, the easiest way is to use Docker:
Copy the .env.example
file to .env
:
cp .env.example .env
Update the values in the .env
file according to your setup.
Repeat steps 1 and 2 for the backend
and inference-gateway
directories:
cd backend-server
cp .env.example .env
# Update .env file with appropriate values
cd inference-gateway
cp .env.example .env
# Update .env file with appropriate values
Run the following command in your terminal:
docker compose up -d
Access the application at http://localhost:3000
NOTE: Local deployment will only set up the database, backend, and frontend servers. Model deployment and finetuning will take place in Kubernetes only.
SimpliML's infrastructure is built on three major components:
The backend is the core API server that handles requests, processes data, and manages the interaction between various components. Built-in pure TypeScript, the backend is designed for performance, scalability, and ease of maintenance.
backend-server
Follow the README.md in the backend-server
directory for detailed setup instructions from the source.
The frontend is a web application built with Next.js, providing an intuitive and interactive user interface for managing models, prompts, and other features of SimpliML.
frontend
Follow the README.md in the frontend
directory for detailed setup instructions from the source.
The inference gateway is responsible for managing inference requests, providing a scalable and efficient way to handle model predictions. Built-in pure TypeScript, it ensures high performance and reliability.
inference-gateway
Follow the README.md in the inference-gateway
directory for detailed setup instructions from the source.
For users looking for a managed solution, SimpliML offers a Managed SaaS version. With the managed version, you can:
Learn more and sign up for SimpliML here.
SimpliML is designed to support multi-cloud deployments using Kubernetes. Kubernetes provides the necessary orchestration and scaling capabilities, allowing you to seamlessly deploy SimpliML across various cloud providers. Whether you're using AWS, Google Cloud Platform (GCP), Azure, or any other cloud service, SimpliML's infrastructure can be easily configured to suit your needs.
Detailed documentation is available to help you navigate and utilize all the features of SimpliML. Please refer to the Documentation for more information.
We welcome contributions from the community! If you want to contribute to SimpliML, please read our Contributing Guide to get started.
SimpliML is licensed under the Apache 2.0 License. You are free to use, modify, and distribute this software according to the terms of the license.
Thank you for choosing SimpliML! We hope you find it valuable in your machine-learning journey. If you have any questions or feedback, please don't hesitate to reach out at support@simpliml.com.