Unsloth AI· @UnslothAI · X·· 2 小时前精选AI 评分66
AI 导读
Unsloth 发布教程与开源仓库，可将 Qwen3.8、Gemma 4 等 LLM 微调为输出选项概率的决策模型，Qwen3.5 0.8B 在 3 个决策基准上的合计准确率从 20.7% 提到 74.3%，仅需 4GB 显存。
推荐理由
原文给出完整微调方法和前后准确率对比，读者可在 4GB 显存上复现把 LLM 转成决策模型的流程。
正文 · 原文
You can now train your own Decision model like Jev locally!
We increased Qwen3.5 0.8B’s aggregate accuracy from 20.7% to 74.3% across 3 decision benchmarks - on just 4GB VRAM.
Turn any LLM like Qwen3.8, Gemma 4 into decision models with our open-source Unsloth repo.
We fine-tuned with a Clef head using Unsloth and LoRA (r=64) for one epoch, increasing downstream accuracy from 30–37% to 78%.
GitHub: https://github.com/unslothai/unsloth
Guide and Notebooks: https://unsloth.ai/docs/basics/train-your-own-decision-model-with-unsloth
来源：Unsloth AI · x.com