spheronFdn / sos-ai-bounty

Spheron Open Source AI Bounty
https://spheron.network
MIT License
52 stars 5 forks source link

MAHA AI HUB Decentralized AI Marketplace within Telegram Mini Apps #8

Open DeImOs-Sj opened 2 weeks ago

DeImOs-Sj commented 2 weeks ago

MAHA AI HUB Decentralized AI Marketplace within Telegram Mini Apps

Submitter Information

Reward Address

0xaE25777C98EE8a66b6A9D5243f3a92455A810ecF

Project Plan

We are working on building the world's first decentralized AI marketplace within Telegram mini apps. This platform empowers users and developers to seamlessly access, trade, and deploy AI models in a secure, transparent, and community-driven environment.

Repository

Github Org: AI Inference Org

Demo

Setup Instructions

Prerequisites

Backend Setup

  1. Run Docker:

    • Navigate to the project directory.
    • Run the following command to start the necessary Docker containers:
      docker-compose up
  2. Run Migrations:

    • After Docker containers are up and running, apply Prisma migrations to set up the database schema:
      npx prisma migrate dev
  3. Start the Backend Server:

    • Once migrations are applied, start the backend server:
      npm run dev

Frontend Setup

  1. Install Dependencies:

    • In the project directory, install the required dependencies:
      npm install
  2. Start the Frontend Server:

    • Run the following command to start the frontend development server:
      npm run dev

      Spheron Infrastructure Usage

      We have utilized Spheron's infrastructure to deploy the AI model for our decentralized marketplace. By leveraging Spheron's robust and scalable deployment platform, we ensured that our AI model is accessible and performant within the Telegram mini apps environment.

For deployment, we used a Docker setup to containerize the application, and the following docker-compose.yml file was used to streamline the deployment process on Spheron:

You can refer to this docker image :- docker compose yml file

Docker image repository = Docker image repo link

DeImOs-Sj commented 2 weeks ago

Commit history trail image

izrake commented 2 weeks ago

Hey, Thank you for creating the issue again.

We are still not able to understand where you are using GPUs.

YAML attached is for the CPUs and models which are deploying it the raw one without any optimization into it. The steps of deploying on Spheron is also not present.