Patrick Devine

It should be working. Here's a session with kimi.

Behold! Server grade RTX 6000 w/ a 3D printed shroud to strap on a cooling fan. This seems like a bad idea, but honestly running great so far. I'm wondering if there are good liquid cooling solutions though.

It's not as fancy as what @ivanfioravanti cooked up with the pagoda demo, but this was a single line prompt to make minecraft in godot using Kimi K3 on @ollama cloud.

19 hours in and still 5 hours to go.

Here's GLM 5.2 for me (through ollama / Hermes)

presense_penalty is set to 1.5, **but** it's not recommended for coding use cases, only chat.

The `coding` tags just have the recommended hyperparameters set for coding/agentic use. They share the same weights (it doesn't take up extra disk space).

Yes! You can find the MLX model tags on ollama.com

Yep. Also we changed ollama.com to show if a model is for MLX. I'm still looking to make this easier though so you don't have to remember all of these long tag names.

it's not in the UI yet, but does work in the CLI. Here's z-image-turbo running on the MLX backend in Ollama in an iterm2 terminal running on macOS:

1-shot prompt Asteroids game w/ GLM-4.7-Flash on the experimental MLX backend for @ollama. This was w/ 8-bit affine quantization and a MBP.

Seems fine w/ gpt-oss.