TRUSTEE-Qwen3-8B

arXiv GitHub License

Introduction

TRUSTEE (Tool Learning Using Simulated Environments) is a cost-friendly online tool learning approach presented in the paper Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Model.

TRUSTEE democratizes tool learning by simulating environments by local open-source LMs, requiring no stablished data, executable tools, or verifiable environments from human experts or advanced proprietary LMs.

Citation

If you find our work useful, feel free to cite our paper:

@misc{tang2026democratizingtoollearningenvironments,
      title={Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Model}, 
      author={Chenming Tang and Hsiu-Yuan Huang and Weijie Liu and Junqiang Zheng and Saiyong Yang and Yunfang Wu},
      year={2026},
      eprint={2604.17739},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2604.17739}, 
}
Downloads last month
16
Safetensors
Model size
8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for JamyDohrn/TRUSTEE-Qwen3-8B

Finetuned
Qwen/Qwen3-8B
Finetuned
(2171)
this model

Collection including JamyDohrn/TRUSTEE-Qwen3-8B

Paper for JamyDohrn/TRUSTEE-Qwen3-8B