How to run Qwen 3.6 locally with Ollama
Short answer: Install Ollama, then run "ollama run qwen3.6". The default qwen3.6 download is about 23-24 GB. For the 27B dense version use "ollama run qwen3.6:27b". Type /bye to exit.
ollama pull qwen3.6 # download (~23-24 GB)
ollama run qwen3.6 # chat in the terminal
ollama run qwen3.6:27b # the 27B dense variant
ollama list # what's installedUse it from other apps
Ollama serves an API on your machine (localhost). Agents like OpenClaw and Hermes Agent can point at a local endpoint, so nothing leaves your computer.
Not enough memory?
Try a smaller quantization tag listed on the Ollama library page, or a smaller model. Close other heavy apps. For fully offline servers, use llama.cpp.
Related questions
Can I use LM Studio instead?
Yes. LM Studio lists Qwen3.6 in its model catalog; download it there and chat in the app.
Where these facts come from
- Ollama library: qwen3.6: pull/run commands and download sizes
- Ollama quickstart: install and first run
- Qwen3.6-27B on Hugging Face: official model card
- LM Studio: Qwen3.6: run Qwen3.6 in LM Studio
Checked Oct 29, 2026. Plans and prices change, so confirm on the official page before you pay.
More answers: all questions →