ArkorAlpha

Find the smallest Gemma 4 that does your job, and train it to do it perfectly.

Cheaper to serve. Faster to run. Small enough to run on-device, even offline.
Coming soon

Two stages: find, then distill.

Stage 01

Find the smallest sufficient model

Start at the largest Gemma 4 size and step down through smaller ones. The pipeline runs your task against each and finds the smallest model that already hits ~80% satisfaction.

Stage 02

Distill it to 100% on your task

Take that small model and train it on outputs from the larger teacher until it stably hits 100% on your task. The result is a small, fast model that behaves the way you need.

Start at 31bTry smaller sizes80% baseline foundDistill on that baseline100% stable output

A smaller model is a different product.

Cheaper to serve

Fewer parameters means fewer GPU cycles per request. Your inference bill scales with model size, not just usage.

Faster to run

Lower latency for the same task. Especially noticeable in interactive product flows where each call is on the critical path.

Runs anywhere

Small enough for laptops, phones, and WebGPU. Run it inside the browser, or offline on-device, when that matters.

Gemma 4 family.

Distil currently supports the Gemma 4 family of open-weight models. Other families will be added as the workflow proves out.

ModelParametersRole
Gemma 4 31b31B (dense)Teacher
Gemma 4 26b A4b26B MoE, 4B activeTeacher
Gemma 4 E4B4B effectiveTeacher / Student
Gemma 4 E2B2B effectiveStudent

More from the toolkit.