huggingface/trl
trl
Train transformer language models with reinforcement learning.
Builder
HuggingFace
huggingface • ai-lab
Stars
17,894
Using upstream star count
Forks
2,604
Using upstream fork count
Open Issues
0
Activity Score
0/100
94 commits in 30d
Created
Mar 27, 2020
Project creation date
README Summary
TRL (Transformer Reinforcement Learning) is a library for training transformer language models using reinforcement learning techniques. It provides tools and algorithms for fine-tuning language models with human feedback, including PPO (Proximal Policy Optimization) and other RL methods. The library integrates seamlessly with Hugging Face's transformers ecosystem for scalable RL training of language models.
AI Dev Skills
Unmapped
Tags
Taxonomy
Deployment Context
Modalities
Skill Areas
Recent Activity
Updated 27 days ago
7 Days
4
30 Days
94
90 Days
416
Quality
production- Quality
- high
- Maturity
- production
Categories
PM Skills
Languages
Timeline
- Project created
- Mar 27, 2020
- Forked
- Mar 13, 2026
- Your last push
- 27 days ago
- Upstream last push
- 6 days ago
- Tracked since
- Mar 17, 2026
Similar Repos
pgvector cosine similarity · $0
Loading…