When LLM Agents Meet Reinforcement Learning
Section: Base Framework · ByteDance
FreshScore 89
Entry
verl
Appears in 5 awesome lists
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 /…
Open github.comvolcengine/verl
Found in these lists
Awesome Agent Evals
Section: 5e · RL-environment / verifiable-reward toolkits (eval ⇄ training) · https://github.com/volcengine/verl (now verl-project/verl) — de-facto industry RLVR trainer (PPO/GRPO). ~22k .
FreshScore 88
Awesome LLM Resources
Section: 微调 Fine-Tuning · Volcano Engine Reinforcement Learning for LLM.
FreshScore 90
Awesome Open Source AI
Section: 7. Training & Fine-tuning Ecosystem · Volcano Engine Reinforcement Learning for LLMs with PPO, GRPO, REINFORCE++, DAPO (EuroSys 2025).
FreshScore 88
Awesome Prompts
Section: Agent Training (RL Post-Training) · 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 /…
FreshScore 96
Often listed next to it
TRL
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…
In 10 listsDetails
unsloth
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
In 8 listsDetails
slime
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)…
In 6 listsDetails
OpenRLHF
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 /…
In 6 listsDetails
Ludwig
Low-code framework for building custom LLMs and deep neural networks. Declarative YAML configuration for training state-of-the-art models with PEFT/LoRA, 4-bit quantization, distributed training via Hugging Face Accelerate, and native Kubernetes support. Linux Foundation AI project. Apache 2.0…
In 6 listsDetails
Kiln
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
In 5 listsDetails
NeMo-RL
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…
In 5 listsDetails
prime-rl
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…
In 5 listsDetails