{"ID":2844610,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.05951","arxiv_id":"2511.05951","title":"Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling","abstract":"Despite the proliferation of powerful agentic models, the lack of critical post-training details hinders the development of strong counterparts in the open-source community. In this study, we present a comprehensive and fully open-source pipeline for training a high-performance agentic model for interacting with external tools and environments, named Klear-Qwen3-AgentForge, starting from the Qwen3-8B base model. We design effective supervised fine-tuning (SFT) with synthetic data followed by multi-turn reinforcement learning (RL) to unlock the potential for multiple diverse agentic tasks. We perform exclusive experiments on various agentic benchmarks in both tool use and coding domains. Klear-Qwen3-AgentForge-8B achieves state-of-the-art performance among LLMs of similar size and remains competitive with significantly larger models.","short_abstract":"Despite the proliferation of powerful agentic models, the lack of critical post-training details hinders the development of strong counterparts in the open-source community. In this study, we present a comprehensive and fully open-source pipeline for training a high-performance agentic model for interacting with extern...","url_abs":"https://arxiv.org/abs/2511.05951","url_pdf":"https://arxiv.org/pdf/2511.05951v1","authors":"[\"Qi Wang\",\"Hongzhi Zhang\",\"Jia Fu\",\"Kai Fu\",\"Yahui Liu\",\"Tinghai Zhang\",\"Chenxi Sun\",\"Gangwei Jiang\",\"Jingyi Tang\",\"Xingguang Ji\",\"Yang Yue\",\"Jingyuan Zhang\",\"Fuzheng Zhang\",\"Kun Gai\",\"Guorui Zhou\"]","published":"2025-11-08T09:47:27Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\"]","has_code":false}