This repository contains Terraform configuration for deploying 371GPT infrastructure on OVH Cloud.
371GPT is a scalable, ethical AI orchestration system for non-coders based on the DSF (Discover, Space, Flow) model. This system enables non-technical users to harness the power of AI agents through intuitive interfaces and automated deployment.
This infrastructure repository is a component of the larger 371GPT system, focusing specifically on the cloud infrastructure deployment using Terraform.
- Accessibility: No-code interfaces for system configuration and management
- Scalability: Microservice architecture with dynamic agent creation
- Ethics: Built-in guardrails and monitoring for responsible AI use
- Robustness: Self-healing error handling and comprehensive logging
- 371GPT Core: Main AI orchestration service with multiple specialized agents
- XPipe Server: Data pipeline management for knowledge processing
- Supabase: Database with vector capabilities (replacing PostgreSQL + MongoDB)
- Kespa: Automation service for workflow orchestration
- Infrastructure Architecture - Detailed explanation of infrastructure components
- Deployment Guide - Step-by-step instructions for deploying infrastructure
- Integration Guide - How to integrate with the main 371GPT system
For more comprehensive information about the 371GPT system:
- Terraform (v1.0.0+)
- OVH Cloud account
- OpenStack credentials for OVH
- SSH key pair configured in OVH Cloud
-
Clone this repository:
git clone https://github.com/ab1355/371gpt-infrastructure.git cd 371gpt-infrastructure
-
Copy the example variables file and add your credentials:
cp terraform.tfvars.example terraform.tfvars # Edit terraform.tfvars with your credentials
-
Initialize Terraform:
terraform init
-
Preview the changes:
terraform plan
-
Apply the configuration:
terraform apply
For Windows, run the provided setup script:
setup-env.bat
For Linux/Mac, source the OpenStack RC file:
source openrc.sh
The infrastructure includes:
- Dedicated security groups with properly configured firewall rules
- Internal network for communication between services
- Limited external access to necessary ports only
- Role-based access control
- Comprehensive audit logging
- GPT Core: s1-8 instance (8 vCPUs, 16GB RAM)
- Supabase: s1-8 instance with 200GB storage
- XPipe: s1-4 instance (4 vCPUs, 8GB RAM)
- Kespa: s1-4 instance (4 vCPUs, 8GB RAM) with 50GB storage
To update the infrastructure:
terraform plan
terraform apply
To tear down the infrastructure:
terraform destroy
This infrastructure repository provides the cloud resources needed to run the complete 371GPT system. After deploying this infrastructure, you'll need to:
- Deploy the application services defined in the main 371GPT repository
- Configure the connections between services as defined in the architecture documentation
- Set up the proper networking and security rules
For more comprehensive information about the 371GPT system:
For information on implementing this system for business purposes, refer to:
Access IP addresses for each service will be displayed in the outputs after successful deployment.