Kev-0.5B: A Tiny Open-Source Decision Model to Train on a MacBook
Melvin Vivas · X post · 2026-09-19 · Open on X
Topics: Fine-tuning & Model Customization · Level: intermediate
Summary
Shares Kev-0.5B, a tiny open-source decision model similar to Jev, with a TypeSafe-compatible API. The quoted post says it is built on Qwen2.5-0.5B, and you can train and run it on a MacBook Pro. The model card and weights are on GitHub.
Key points
- Kev-0.5B is a small decision model, about 0.5B parameters.
- The quoted post says the base is Qwen2.5-0.5B, but the GitHub repo title says Qwen3.5/3.8. Check the repo.
- Comes with a TypeSafe-compatible API.
- Small enough to train and run on a MacBook Pro.
- Model card and weights are on GitHub (jaredpalmer/kev).
Resources mentioned
- jaredpalmer/kev (GitHub) · repo · github.com · free
Family of Jev-like decision models built on Qwen that you can train and run yourself. - Qwen2.5-0.5B · tool · huggingface.co · free
Tiny open-weight Qwen model, named as the base for Kev-0.5B.
Try this
- Train your own model; try Kev-0.5B from the GitHub repo.
- Train and run a tiny decision model locally on a MacBook.
More in Fine-tuning & Model Customization
- Hugging Face AutoTrain: a no-code fine-tuning tool
- LoRA Fine-Tune Qwen3.5-2B on Your Tweets with Unsloth Studio
- Fine-Tune ModernBERT-base as a Task-Routing Classifier
- Train ModernBERT as a Prompt Router Between Two Models (Colab Notebook)
- Fine-tuning ModernBERT-base as a task router (quote post)
- Fine-tuning ModernBERT-base to route tasks between two models