- update the fine-tuning scripts; - add new models: Qwen;[QwQ 32B](https://qwenlm.github.io/blog/qwq-32b/); @LChenGit pick models