Home-Qwen3.5-0.8B-v0

Home-Qwen3.5-0.8B-v0 is an experimental finetune for use in Home Assistant.

Why v0?

This is my first finetune ever, and frankly I would not be surprised if I broke something or did it wrong to begin with.

If It Just Works, Great :D! If you recognize an issue, please bring it to my attention.

---

This qwen3_5 model was trained 2x faster with Unsloth and Huggingface's TRL library.