Trinity-RFT and LightRFT

These are competitors in the RFT space, both targeting LLM fine-tuning through reinforcement learning but with different design philosophies—Trinity-RFT emphasizing general-purpose flexibility and scalability while LightRFT prioritizes lightweight efficiency and multimodal reward modeling.

Trinity-RFT
69
Established
LightRFT
51
Established
Maintenance 17/25
Adoption 10/25
Maturity 25/25
Community 17/25
Maintenance 10/25
Adoption 10/25
Maturity 22/25
Community 9/25
Stars: 557
Forks: 55
Downloads:
Commits (30d): 9
Language: Python
License: Apache-2.0
Stars: 208
Forks: 10
Downloads:
Commits (30d): 0
Language: Python
License: Apache-2.0
No risk flags
No risk flags

About Trinity-RFT

agentscope-ai/Trinity-RFT

Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).

This helps developers fine-tune large language models (LLMs) using reinforcement learning. You provide an existing LLM and define an environment for it to interact with, and this framework helps train the model to perform better at specific tasks. It is for AI developers, machine learning engineers, and researchers who want to improve the performance of their LLMs.

LLM fine-tuning agent training reinforcement learning AI model development natural language processing

About LightRFT

opendilab/LightRFT

LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework

This framework helps AI practitioners improve the performance and behavior of Large Language Models (LLMs) and Vision-Language Models (VLMs). You feed in a pre-trained language or vision-language model along with human feedback or a reward model, and it outputs a fine-tuned model that better aligns with desired outcomes, like generating more accurate text or understanding multimodal data. It's designed for machine learning engineers and researchers working with advanced AI models.

large-language-models vision-language-models reinforcement-learning model-fine-tuning multimodal-ai

Scores updated daily from GitHub, PyPI, and npm data. How scores work