When LLM Agents Meet Reinforcement Learning
Section: Base Framework · HuggingFace
Entry
Appears in 10 awesome lists
Hugging Face's post-training library — the reference implementation of GRPO (the algorithm DeepSeek popularized) and the most-used single-GPU-to-cluster starting point for RL post-training; 75+ trainers across SFT / DPO / KTO / PPO / ORPO / CPO and more, with vLLM generation, PEFT/LoRA, and AMD…
Section: Base Framework · HuggingFace
Section: Foundation Model Fine Tuning · Train transformer language models with reinforcement learning.
Section: Training and Fine-tuning · train transformer language models with reinforcement learning
Section: Instruction Tuning and Preference Optimization · reference library for SFT, DPO, GRPO, and RLHF.
Section: 7. Training & Fine-tuning Ecosystem · Official library for RLHF, SFT, DPO, ORPO.
Section: Industry Strength Reinforcement Learning · Train transformer language models with reinforcement learning.
Section: Agent Training (RL Post-Training) · Hugging Face's post-training library — the reference implementation of GRPO (the algorithm DeepSeek popularized) and the most-used single-GPU-to-cluster starting point for RL post-training; 75+ trainers across SFT / DPO / KTO / PPO / ORPO / CPO and more, with vLLM generation, PEFT/LoRA, and AMD…
Section: Other · Train transformer language models with reinforcement learning.
Section: Graph Machine Learning · Train transformer language models with reinforcement learning.
Section: AI and Agents · A library for post-training transformer language models with SFT, DPO, GRPO, and other trainers.
Parameter-Efficient Fine-Tuning (PEFT) methods enable efficient adaptation of pre-trained language models (PLMs) to various downstream applications without fine-tuning all the model's parameters.
Zhipu AI / Tsinghua's (THUDM) RL scaling framework — the production trainer behind the GLM series from GLM-4.5 to GLM-5.3-Flash; connects Megatron training with SGLang rollout through a unified train–rollout–buffer architecture (parameters sync, router-based generation, prompt/verifier management)…
The Ray+vLLM pioneer of open RLHF/RL post-training — the first framework built around a fully decoupled Ray-based distributed architecture (vLLM generation + DeepSpeed ZeRO/FSDP training, placement-colocated hybrid engine with sleep-mode weight sharing between actor/critic/reward/reference); PPO /…
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
NVIDIA's scalable RL post-training library (successor to NeMo Aligner, re-architected with the NeMo Framework) — the megatron-core-backed option in the open stack: GRPO / GSPO / DAPO / PPO / DPO / SFT(+LoRA) / on-policy distillation / GDPO with training via PyTorch DTensor (FSDP2) or Megatron Core…
Prime Intellect's fully asynchronous RL framework — "agentic RL training at scale" (the INTELLECT-3 report) — built to train 1T+ parameter MoE models on 1,000+ GPUs through three abstractions: a lightweight CPU Orchestrator (dataflow/scheduling), an FSDP2 Trainer, and vLLM inference as…
The most widely adopted open-source RL post-training framework for LLMs (ByteDance Seed; HybridFlow architecture, EuroSys 2025) — a flexible single-controller/dataflow abstraction that hybrid-orchestrates rollout generation (vLLM/SGLang) and training (FSDP/Megatron) in one program; PPO / GRPO /…
a Python library for using and training embedding and reranker models for applications like retrieval augmented generation, semantic search, and more