TRL
JSON →Hugging Face library for post-training LLMs: SFT, DPO, GRPO, PPO, reward modeling. Current version is 0.29.1 (Mar 2026). Requires Python >=3.10. Extremely high API churn — major parameter renames across versions. tokenizer= renamed to processing_class= in 0.12. Still pre-1.0 (Development Status: Pre-Alpha).
Traffic · last 30 days ↑67% vs prev 7d
total hits 38
actors 8 distinct systems
last hit 8d ago human
top countries 🇺🇸 United States · 🇸🇬 Singapore · 🇨🇦 Canada · VE · VN
API endpoints
full doc /v1/registry/trl
install /v1/registry/trl/install
compatibility /v1/registry/trl/compatibility