Plans

Odysseus VPS Plans Coming Soon

We are preparing VPS plans with Odysseus VPS pre-installed. Below are the resources the application needs to run smoothly.
Minimum
  • Processor 2 vCPU
  • Memory (RAM) 4 GB
  • NVMe storage 30 GB

Together with the vector database and search service; for use with cloud APIs.

Recommended
  • Processor 4 vCPU
  • Memory (RAM) 8 GB
  • NVMe storage 80 GB

For deep research, document work and several modules. Our servers have no GPU.

Get notified when plans launch

Contact our team to learn more about Odysseus VPS plans or to plan the right setup for your needs today.

Contact Us Browse Linux VPS plans
Use Cases

What Can You Do with Odysseus?

Bring your AI work, from chat and research to notes and images, into one workspace.

Chat

Chat with cloud and local models from one interface and switch between them.

Deep research

Have a topic researched in multiple steps and get comprehensive reports with sources.

Autonomous agents

Hand multi-step tasks to agents and follow their progress from the workspace.

Notes and productivity

Manage notes, calendar and email in a single AI-assisted workspace.

Image generation

Generate images with the models you connect and keep them in your workspace.

Model guide

Discover models that suit your hardware and needs with the Cookbook section.

Why TRVDS for Odysseus VPS?

Your application runs in our data center in Bursa, Turkey, on resources reserved for you alone.

Pre-installed

The application is installed automatically after your server is created. No need to deal with Docker or dependencies.

Dedicated resources

Intel Xeon Gold processors, DDR4 RAM and NVMe storage. CPU and memory are not shared with other customers.

DDoS protection and unlimited traffic

DDoS protection with 40 Gbps capacity and unlimited traffic are included in every VPS plan.

Full control

With root access you can update the application, add a domain and SSL, and install other services alongside it.

What Is Odysseus?

Odysseus is an open-source, self-hosted AI workspace developed by Felix Kjellberg (PewDiePie) and released in May 2026. It combines chat, autonomous agents, Deep Research, productivity tools such as notes, calendar and email, and image generation in a single interface.

It works with local model servers such as Ollama, llama.cpp and vLLM, as well as OpenRouter and OpenAI-compatible APIs. Its “Cookbook” section recommends models that suit your hardware.

Running It on Your Own Server

Odysseus’s modules run on your server and connect only to the endpoints you configure. When you connect a cloud model, your requests go to that provider, while your notes, research results and chat history stay on your server.

Our VPS plans do not include a graphics card (GPU), so we recommend using Odysseus with OpenRouter or an OpenAI-compatible API. Large local models require GPU hardware with more than 16 GB of RAM.

Maturity, Security and Resources

Odysseus is a new and fast-moving project; it is updated often and some features are still maturing. The agents’ shell tool does not run in a sandbox but with the permissions of the user running the application. We therefore recommend not exposing the interface to the internet without a password and SSL, and not keeping other critical data on the server. If stability is your priority, consider Open WebUI or LibreChat.

Odysseus is licensed under AGPL-3.0. Because it runs together with components such as a vector database and a search service, we recommend at least 2 vCPU and 4 GB RAM, and 4 vCPU and 8 GB RAM for deep research and document work.

Other Server Options

Looking for a different server?

Compare our other server options by operating system, control panel or resources.
FAQ

Frequently Asked Questions About Odysseus VPS

Check the most frequently asked questions and find answers instantly.

Odysseus is an open-source, self-hosted AI workspace that combines chat, AI agents, deep research, notes and image generation in one interface. It was developed by Felix Kjellberg (PewDiePie).

Yes. Odysseus is licensed under AGPL-3.0 and free to use on your own server. Cloud model usage is billed by the provider.

Yes. When used with cloud APIs, no GPU is needed. Running large local models requires GPU hardware with plenty of RAM; our VPS plans do not include a GPU.

Together with its vector database and search service it needs at least 2 vCPU and 4 GB RAM. For deep research and document work we recommend 4 vCPU and 8 GB RAM.

Odysseus is a new project, and the agents’ shell tool does not run in a sandbox. We recommend not exposing the interface to the internet without a password and SSL, and not keeping other critical data on the server.

Open WebUI is a mature, stable chat interface. Odysseus adds agents, deep research, notes and calendar tools to chat, and is a newer, fast-evolving workspace.