{"id":1888,"date":"2026-08-19T13:09:19","date_gmt":"2026-08-19T13:09:19","guid":{"rendered":"https:\/\/blog.openzeka.com\/en\/?p=1888"},"modified":"2026-08-20T05:51:09","modified_gmt":"2026-08-20T05:51:09","slug":"local-ai-workspace-with-odysseus-on-dgx-spark","status":"publish","type":"post","link":"https:\/\/blog.openzeka.com\/en\/local-ai-workspace-with-odysseus-on-dgx-spark\/","title":{"rendered":"Local AI Workspace with Odysseus on DGX Spark"},"content":{"rendered":"

In the previous tutorial, we learned how to run the GPT-OSS 120B model on a DGX Spark using sparkrun. There, we used Open WebUI as the web interface. In this tutorial, we will explore a more advanced AI workspace: Odysseus.<\/p>\n

Odysseus is an open-source AI workspace (84.3k stars on GitHub, AGPL-3.0 license). While it provides a chat interface like Open WebUI, it also offers the following features:<\/p>\n

    \n
  • Agents:<\/strong> The model autonomously performs operations such as writing files, running shell commands, and using tools<\/li>\n
  • Deep Research:<\/strong> Multi-step web research, source reading, and report generation<\/li>\n
  • Terminal access:<\/strong> The model runs commands through a sandboxed terminal<\/li>\n
  • Document management:<\/strong> Document upload, editing, and AI-powered analysis<\/li>\n
  • Email:<\/strong> IMAP\/SMTP inbox, triage, and reply drafts<\/li>\n
  • Calendar and Notes:<\/strong> Reminders, to-dos, and scheduled tasks<\/li>\n
  • Code execution:<\/strong> Writing, saving, and testing programs<\/li>\n
  • Web search:<\/strong> Real-time web search with SearXNG integration<\/li>\n<\/ul>\n

    Odysseus can connect to any OpenAI-compatible inference server such as vLLM, Ollama, or llama.cpp, and manage multiple models simultaneously. In this tutorial, we will use GPT-OSS 120B. However, you can use any other model if you prefer.<\/p>\n

    This tutorial consists of three parts:<\/p>\n

      \n
    • Setup:<\/strong> Downloading, configuring, and starting Odysseus<\/li>\n
    • Using Odysseus:<\/strong> Model connection, chat, document analysis, deep research, and agent mode<\/li>\n
    • Shutdown:<\/strong> Stopping services and data persistence<\/li>\n<\/ul>\n

      You can refer to previous tutorials for running the GPT-OSS 120B model (or other models we used in the tutorials). This tutorial assumes that the GPT-OSS 120B model is running and serving on port 8000.<\/p>\n

      \n

      Setup<\/h2>\n

      1. Downloading Odysseus<\/h3>\n

      Odysseus uses a Docker Compose stack with 4 services: Odysseus web interface, ChromaDB (vector database), SearXNG (web search engine), and ntfy (push notifications). First, clone the repository:<\/p>\n<\/div>