LightRFT
by opendilab · GitHub
LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework
⭐ Stars
403
🍴 Forks
11
📜 License
Apache-2.0
Commercial use OK📅 Created
2025
🔄 Last commit
2 mo ago
🏷️ Category
dapo
💻 Language
You maintain this project?
Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.
📈 Star history
2026-07-202026-07-21
📄 About
LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework
❓ Frequently asked questions
What is LightRFT?
LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework
Is LightRFT open source?
LightRFT is an open-source project. It is released under the Apache-2.0 license.
Is LightRFT free?
Yes. LightRFT is free and open source — you can use, modify and self-host it.
🏅 Maintainer of this project?
Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.
[](https://opensourceai.tech/project/opendilab-lightrft.html)
🧬 Related projects🧬 View the DNA map →
ms-swift
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, Intern…
★ 14.9k · deepseek-r1
ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your…
★ 10.5k · agent
VLM-R1
Solve Visual Understanding with Reinforced VLMs
★ 6k · deepseek-r1