AIPython★ 682⑂ 19 forksApache-2.0
baidu-baige/LoongSage
Agentic RL Framework for LLMs
Production-grade framework for training frontier LLMs with agentic reinforcement learning and validated recipes.
// KEY FEATURES
- Implements GRPO, DAPO, and PPO algorithms for efficient LLM training
- Supports distributed training across multi-node GPU clusters with vLLM integration
- Provides validated recipes for math, coding, and reasoning benchmarks
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram