TRL

Train transformer language models with reinforcement learning: SFT, DPO, PPO, and GRPO.

Bintang

19,688

Fork

3,124

Bahasa

Python

Kategori

Training & Fine-tuning

Dapatkan kode

# Clone repositori
git clone https://github.com/huggingface/trl

Apa itu TRL?

Train transformer language models with reinforcement learning: SFT, DPO, PPO, and GRPO.

Apakah TRL aman digunakan?

TRL adalah proyek open source dari huggingface di GitHub. Tinjau kode sumber dan lisensi sebelum menggunakannya di produksi.

Bagaimana memulai dengan TRL?

git clone https://github.com/huggingface/trl

TRL ditulis dalam bahasa apa?

openSource.faqWhatLanguageAnswer

Fine-tune and run LLMs and diffusion models 2x faster with far less memory.

Unified efficient fine-tuning of 100+ LLMs and VLMs with a no-code web UI.

Deep learning optimization library that makes distributed training and inference easy and efficient.