Master Agent Cloud with our step-by-step tutorial, detailed feature walkthrough, and expert tips.
Clone the Agent Cloud repository from GitHub (git clone https://github.com/rnadigital/agentcloud.git) and ensure Docker is installed with at least 16 GB RAM allocated to the Docker engine Run the automated installer with 'chmod +x install.sh && ./install.sh' on Mac or Linux, then follow the interactive prompts to configure your LLM provider (local via Ollama/LM Studio or cloud via OpenAI API key) Open the Agent Cloud web interface at localhost:3000, create your first agent by selecting an LLM model and defining its role and instructions through the visual agent builder Connect your first data source using the Data Sources panel — choose from 260+ connectors (e.g., upload a PDF, connect a PostgreSQL database, or link a Confluence workspace) and configure the sync schedule Start a new chat session with your configured agent to test RAG responses against your connected data, then iterate on chunking settings and agent instructions to optimize answer quality
💡 Quick Start: Follow these 1 steps in order to get up and running with Agent Cloud quickly.
Explore the key features that make Agent Cloud powerful for ai memory & search workflows.
Agent Cloud requires a machine with at least 16 GB of RAM for Docker-based deployment. A base MacBook Air M1/M2 with 8 GB RAM is insufficient as the Airbyte integration requires significant resources. If running local LLMs via Ollama or LM Studio alongside Agent Cloud, additional RAM is recommended. Non-Docker deployments may work with 8 GB RAM but are harder to configure.
Yes. By using local LLM providers like Ollama or LM Studio and connecting only to on-premises data sources, Agent Cloud can operate in a fully air-gapped environment with zero external API calls. This makes it suitable for classified or highly regulated environments where internet connectivity is restricted.
AGPL 3.0 is a copyleft open-source license that allows free use, modification, and deployment. However, if you modify the source code and distribute the software or provide it as a network service to others, you must make your modifications available under the same license. Internal use within your organization does not trigger this requirement.
Agent Cloud provides complete data sovereignty (your data never leaves your servers), supports 260+ data source integrations vs GPTs' limited file upload approach, enables multi-agent orchestration for complex workflows, and has no per-token usage fees beyond your own infrastructure costs. The trade-off is that Agent Cloud requires self-hosting and technical setup, while custom GPTs are instantly available but route all data through OpenAI's servers.
Agent Cloud natively supports Qdrant (included in the Docker deployment) and Pinecone. The platform's Rust-based vector proxy provides high-performance communication with these databases for fast similarity search across large document collections.
Yes. While initial deployment requires Docker and DevOps knowledge, the day-to-day operation of Agent Cloud uses an intuitive web-based GUI. Non-technical team members can create agents, connect data sources, manage conversations, and configure workflows through the visual interface without touching the command line.
The community edition is free to download and run. Your real costs are infrastructure and LLM API fees. A typical small-team deployment on AWS (m5.xlarge instance, EBS storage, and OpenAI API usage for ~50 users) runs roughly $200–$500/month all-in. Managed Cloud pricing is usage-based and starts in the $500–$2,000/month range depending on cluster size and connector volume — contact RNA Digital's sales team for an exact quote. Enterprise contracts are annual and typically range from $25,000 to $100,000+ per year based on deployment model, seat count, and support tier. For budget planning, the self-hosted path is significantly cheaper than comparable managed platforms like Dify Cloud or Botpress Enterprise, but requires DevOps investment.
Now that you know how to use Agent Cloud, it's time to put this knowledge into practice.
Sign up and follow the tutorial steps
Check pros, cons, and user feedback
See how it stacks against alternatives
Follow our tutorial and master this powerful ai memory & search tool in minutes.
Tutorial updated March 2026