Skip to content
This repository was archived by the owner on Nov 9, 2026. It is now read-only.

v0.3.1

Choose a tag to compare

@speedyrulz speedyrulz released this 28 Oct 22:54
· 5 commits to master since this release

Pick which draft (MTP) model RulZ uses, from your phone.

Needs RulZ Coding Agent 0.9.2 and RulZ LLM Suite 0.3.9 on the computer for the draft model choice; everything else works as in 0.3.0.

What's new

  • Settings > AI model > Model settings > How it loads > Draft model: Automatic, the MTP layers built into the model (many Qwen 3.5+ "MTP" models), or any draft file in RulZ's models folder, such as another quantization's mtp-*.gguf.
  • The draft switch says what Automatic uses, and Run the draft on is hidden for built-in layers, which run on the model's own weights.
  • A reply that was only the model thinking no longer shows as an empty row.

Install

Download RulZ-Agent-0.3.1.apk on the phone, open it, and allow installing from that source. It installs over 0.3.0 and keeps your paired computers.

This is a debug-signed build (Android 7.0 and newer).