Fine-tune LLMs from one YAML config - train 8B models on 4GB laptop GPUs with layer streaming
A modular full-stack reinforcement learning library for training and fine-tuning LLMs with advanced agent capabilities
Generate, Clean, and Prepare LLM Data with AI-Powered Operators and Pipelines
AI-optimized modular instructions to help LLMs master Android development best practices
Async reinforcement learning framework for training 1T+ parameter agentic AI models at scale
RL environments and evaluation framework for training and testing LLMs
Fine-tune 600+ LLMs and 300+ MLLMs with PEFT, full-parameter training, and advanced RL algorithms.
Open-source framework for training and researching foundation models with full reproducibility.
Train multi-step AI agents with reinforcement learning using GRPO
