Models

What we publish

Requants and hardware-specific builds of the models we run ourselves, published on HuggingFace. One model family so far; more will land here as we quantize them.

Illustration of a llama peeking out of a small server box, with a glowing halo above its head

Qwen3.8-Flash-Next

Seven mainline-compatible tiers plus a Strix Halo build, each benchmarked against the leading community builds at matching sizes. Our smallest tier beats the field in its class.