Q

qwen

🔧🔗https://github.com/modelscope/ms-swift

SWIFT (Scalable lightWeight Infrastructure for Fine-Tuning) Use PEFT or Full-parameter to finetune 400+ LLMs or 100+ MLLMs.

🔧🔗https://github.com/om-ai-lab/VLM-R1 Solve Visual Understanding with Reinforced VLMs

🔧🔗https://github.com/QwenLM/online_merging_optimizers Implementations of online merging optimizers proposed by Online Merging Optimizers for Boosting Rewards and Mitigating Tax in Alignment