Skip to content
Awesomelist
Type at least 2 letters

Search entries, descriptions and list names. Press Esc to close.

Arrow keys move, Enter opensFull search page

Entry

NeMo-RL

Appears in 5 awesome lists

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…

Open github.comnvidia-nemo/rl

Found in these lists

When LLM Agents Meet Reinforcement Learning

Section: Base Framework · NVIDIA

FreshScore 89

Awesome local LLM

Section: Training and Fine-tuning · scalable toolkit for efficient model reinforcement

FreshScore 90

Awesome Open Source AI

Section: 7. Training & Fine-tuning Ecosystem · Scalable toolkit for efficient model reinforcement with DTensor and Megatron backends.

FreshScore 88

Awesome Production Machine Learning

Section: Industry Strength Reinforcement Learning · NeMo RL is a scalable and efficient post-training library for reinforcement learning on LLMs, supporting algorithms such as GRPO and PPO.

FreshScore 99

Awesome Prompts

Section: Agent Training (RL Post-Training) · 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…

FreshScore 96

Colossal-AI

(from Hpcaitech) - A Unified Deep Learning System for Large-Scale Parallel Training (1D, 2D, 2.5D, 3D and sequence parallelism, and ZeRO protocol).

In 14 listsDetails

Ray

A fast and simple framework for building and running distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library. ray.io

In 13 listsDetails

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

PEFT

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.

In 7 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

LeRobot

Making AI for robotics more accessible with end-to-end learning. State-of-the-art approaches for imitation learning and reinforcement learning with pretrained models, datasets, and simulated environments. Apache 2.0 licensed.

In 6 listsDetails