Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pan, Xuchen, Chen, Yanxi, Chen, Yushuo, Sun, Yuchang, Chen, Daoyuan, Zhang, Wenhao, Xie, Yuexiang, Huang, Yilun, Zhang, Yilei, Gao, Dawei, Shi, Weijie, Li, Yaliang, Ding, Bolin, Zhou, Jingren
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916975476736000
author Pan, Xuchen
Chen, Yanxi
Chen, Yushuo
Sun, Yuchang
Chen, Daoyuan
Zhang, Wenhao
Xie, Yuexiang
Huang, Yilun
Zhang, Yilei
Gao, Dawei
Shi, Weijie
Li, Yaliang
Ding, Bolin
Zhou, Jingren
author_facet Pan, Xuchen
Chen, Yanxi
Chen, Yushuo
Sun, Yuchang
Chen, Daoyuan
Zhang, Wenhao
Xie, Yuexiang
Huang, Yilun
Zhang, Yilei
Gao, Dawei
Shi, Weijie
Li, Yaliang
Ding, Bolin
Zhou, Jingren
contents Trinity-RFT is a general-purpose, unified and easy-to-use framework designed for reinforcement fine-tuning (RFT) of large language models. It is built with a modular and decoupled design, consisting of (1) an RFT-core that unifies and generalizes synchronous/asynchronous, on-policy/off-policy, and online/offline modes of RFT; (2) seamless integration for agent-environment interaction with high efficiency and robustness; and (3) systematic data pipelines optimized for RFT. Trinity-RFT can be easily adapted for diverse application scenarios, and serves as a unified platform for development and research of advanced reinforcement learning paradigms at both macroscopic and microscopic levels. This technical report outlines the vision, features, design and implementations of Trinity-RFT, accompanied by extensive examples, applications and experiments that demonstrate its functionalities and user-friendliness.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17826
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models
Pan, Xuchen
Chen, Yanxi
Chen, Yushuo
Sun, Yuchang
Chen, Daoyuan
Zhang, Wenhao
Xie, Yuexiang
Huang, Yilun
Zhang, Yilei
Gao, Dawei
Shi, Weijie
Li, Yaliang
Ding, Bolin
Zhou, Jingren
Machine Learning
Computation and Language
Distributed, Parallel, and Cluster Computing
Trinity-RFT is a general-purpose, unified and easy-to-use framework designed for reinforcement fine-tuning (RFT) of large language models. It is built with a modular and decoupled design, consisting of (1) an RFT-core that unifies and generalizes synchronous/asynchronous, on-policy/off-policy, and online/offline modes of RFT; (2) seamless integration for agent-environment interaction with high efficiency and robustness; and (3) systematic data pipelines optimized for RFT. Trinity-RFT can be easily adapted for diverse application scenarios, and serves as a unified platform for development and research of advanced reinforcement learning paradigms at both macroscopic and microscopic levels. This technical report outlines the vision, features, design and implementations of Trinity-RFT, accompanied by extensive examples, applications and experiments that demonstrate its functionalities and user-friendliness.
title Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models
topic Machine Learning
Computation and Language
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2505.17826