Liquid AI Releases Free Tutorials to Fine-Tune Decision Models in 30 Minutes

Liquid AI's DevRel engineer Leonie released two open-source notebooks showing how to fine-tune the d1-3B decision model in about 30 minutes on a free GPU using Unsloth and Hugging Face TRL. A separate team demonstrated that fine-tuning the smaller d1-omni-600M on game (state, input) data lets it outperform much larger models on that task.

2026-10-16 ~ 2026-10-17 · 4 related posts

• Liquid AI tutorial: fine-tune your d1-3B decision model in 30 minutes on a free GPU — helloiamleonie · 2026-10-16
• Fine-tune the d1-3B decision model in 30 minutes on a free-tier GPU — danielhanchen · 2026-10-16
• Fine-tuned 600M decision model crushes bigger models; free 30-min tutorial released — iamrobotbear · 2026-10-17

1 near-duplicate retellings: maximelabonne