The fully open post-training corpus behind SmolLM3-3B, organized into three subsets matching the model's training phases: mid-training (4.8M rows), SFT (a decontaminated mixture of ~24 datasets including OpenThoughts, Tulu 3, OpenHermes, and multilingual data), and preference (447k rows for APO). The successor to SmolTalk — a complete, reproducible recipe for modern small-model post-training including dual reasoning/no-reasoning modes.
| Provider | HuggingFaceTB |
| Category | 指令 / SFT |
| Size | 3 Subsets (Mid 4.8M Rows) |
| License | Apache 2.0 |
| Downloads | n/a |
| Tags | Post-Training, SmolLM3, Reasoning, Multilingual, 2025 |
from datasets import load_dataset
ds = load_dataset("HuggingFaceTB/smoltalk2")
使用 QLoRA(4-bit 基础模型 + LoRA 适配器)微调的预计显存需求(保守默认参数):
| 7B QLoRA | ~6GB VRAM |
| 13B QLoRA | ~10GB VRAM |
微调新手?跟着分步教程走: 一小时微调你的第一个 LLM