Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Qi, Zhang, Hongzhi, Fu, Jia, Fu, Kai, Liu, Yahui, Zhang, Tinghai, Sun, Chenxi, Jiang, Gangwei, Tang, Jingyi, Ji, Xingguang, Yue, Yang, Zhang, Jingyuan, Zhang, Fuzheng, Gai, Kun, Zhou, Guorui
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917069457457152
author Wang, Qi
Zhang, Hongzhi
Fu, Jia
Fu, Kai
Liu, Yahui
Zhang, Tinghai
Sun, Chenxi
Jiang, Gangwei
Tang, Jingyi
Ji, Xingguang
Yue, Yang
Zhang, Jingyuan
Zhang, Fuzheng
Gai, Kun
Zhou, Guorui
author_facet Wang, Qi
Zhang, Hongzhi
Fu, Jia
Fu, Kai
Liu, Yahui
Zhang, Tinghai
Sun, Chenxi
Jiang, Gangwei
Tang, Jingyi
Ji, Xingguang
Yue, Yang
Zhang, Jingyuan
Zhang, Fuzheng
Gai, Kun
Zhou, Guorui
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05951
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling
Wang, Qi
Zhang, Hongzhi
Fu, Jia
Fu, Kai
Liu, Yahui
Zhang, Tinghai
Sun, Chenxi
Jiang, Gangwei
Tang, Jingyi
Ji, Xingguang
Yue, Yang
Zhang, Jingyuan
Zhang, Fuzheng
Gai, Kun
Zhou, Guorui
Artificial Intelligence
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.
title Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling
topic Artificial Intelligence
url https://arxiv.org/abs/2511.05951