Using Codex CLI with gpt-oss:120b on an NVIDIA DGX Spark via Tailscale

This technical blog post details how to configure and use OpenAI's Codex CLI coding agent with the gpt-oss:120b model running locally in Ollama on an NVIDIA DGX Spark. The setup uses a Tailscale network to allow the author to run the Codex CLI from their laptop anywhere in the world against this self-hosted model, demonstrating its use by building a Space Invaders clone.

0 Comments