Alternatives and similar repositories for RLFromScratch

Users that are interested in RLFromScratch are comparing it to the libraries listed below

• McGill-NLP / nano-aha-moment
  Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"
  ☆589Updated 3 months ago
• facebookresearch / PhysicsLM4
  Physics of Language Models: Part 4.2, Canon Layers at Scale where Synthetic Pretraining Resonates in Reality
  ☆316Updated last month
• huggingface / picotron_tutorial
  ☆230Updated 2 months ago
• changjonathanc / flex-nano-vllm
  FlexAttention based, minimal vllm-style inference engine for fast Gemma 2 inference.
  ☆334Updated 3 months ago
• tokenbender / avataRL
  rl from zero pretrain, can it be done? yes.
  ☆286Updated 4 months ago
• wolfecameron / nanoMoE
  An extension of the nanoGPT repository for training small MOE models.
  ☆233Updated 10 months ago
• brendanhogan / DeepSeekRL-Extended
  Exploring Applications of GRPO
  ☆251Updated 5 months ago
• JinjieNi / MegaDLMs
  GPU-optimized framework for training diffusion language models at any scale. The backend of Quokka, Super Data Learners, and OpenMoE 2 tr…
  ☆321Updated 2 months ago
• ServiceNow / PipelineRL
  A scalable asynchronous reinforcement learning implementation with in-flight weight updates.
  ☆361Updated this week
• thinking-machines-lab / batch_invariant_ops
  ☆957Updated 3 months ago
• meta-pytorch / torchforge
  PyTorch-native post-training at scale
  ☆605Updated last week
• fangyuan-ksgk / Tiny-GRPO
  minimal GRPO implementation from scratch
  ☆102Updated 10 months ago
• shangshang-wang / Tina
  Tina: Tiny Reasoning Models via LoRA
  ☆316Updated 4 months ago
• openai / circuit_sparsity
  Open-source release accompanying Gao et al. 2025
  ☆501Updated last month
• stanford-cs336 / assignment5-alignment
  ☆104Updated 6 months ago
• radixark / miles
  Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
  ☆830Updated this week
• microsoft / dion
  Dion optimizer algorithm
  ☆424Updated 3 weeks ago
• ZihanWang314 / CoE
  Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models
  ☆227Updated 3 months ago
• test-time-training / e2e
  Official JAX implementation of End-to-End Test-Time Training for Long Context
  ☆511Updated last week
• NVlabs / RLP
  [ICLR 2026] Official PyTorch Implementation of RLP: Reinforcement as a Pretraining Objective
  ☆231Updated last week
• lmgame-org / GRL
  Multi-Turn RL Training System with AgentTrainer for Language Model Game Reinforcement Learning
  ☆59Updated last month
• open-thought / tiny-grpo
  Minimal hackable GRPO implementation
  ☆321Updated last year
• test-time-training / discover
  ☆394Updated last week
• axon-rl / gem
  A Gym for Agentic LLMs
  ☆439Updated 2 weeks ago
• sail-sg / oat
  🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.
  ☆623Updated last week
• joey00072 / nanoGRPO
  nanoGRPO is a lightweight implementation of Group Relative Policy Optimization (GRPO)
  ☆143Updated 8 months ago
• stanford-cs336 / spring2024-lectures
  ☆412Updated last year
• marin-community / marin
  Open-source framework for the research and development of foundation models.
  ☆742Updated this week
• SakanaAI / RLT
  Training teachers with reinforcement learning able to make LLMs learn how to reason for test time scaling.
  ☆358Updated 7 months ago
• facebookresearch / memory
  Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, spars…
  ☆371Updated last year