
How to use Ollama with Codex CLI: ollama launch codex + local gpt-oss
ollama launch codex, the --oss flag, or a permanent profile, plus the 64k context Codex needs and what gpt-oss 20B vs 120B costs in VRAM at that length.
Aliteq Team · · 6 min
Guides, answers & tools
Run AI models on your own PC: apps, models, VRAM and image generation · page 4