@michabbb@social.vivaldi.net2025-07-13 20:44:49
🤖 Models available: WebSailor-3B on #HuggingFace & WebDancer-QwQ-32B for complex reasoning
🛠️ Complete post-training methodology with DUPO algorithm for efficient agentic RL
https://github.com/Alibaba-NLP/W…