• Skip to primary navigation
  • Skip to content
  • Skip to footer
Ax3l Ax3l

      Nadim-Daniel Ghaznavi

      Creator

      • Hamilton, ON
      • GitHub
      • My Website
      • Email

      Environment Setup

      Ax3l

      • OS Setup
      • Shared LlamaSrv setup
      • Ax3l installation and configuration

      Provision LlamaSrv independently with the Qwen3.5-4B model and its vision projector. Ax3l connects to its shared inference endpoint, normally port 51001; configure llamasrv_url in /opt/prod/ax3l/conf/ax3l.json when the service runs on another host. Ax3l itself needs no model files or GPU setup.


      Back

      • Feed
      © 2026 Ax3l. Powered by Jekyll & Minimal Mistakes.