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"
  ☆535Updated 2 months ago
• shangshang-wang / Tina
  Tina: Tiny Reasoning Models via LoRA
  ☆284Updated 2 weeks ago
• tokenbender / avataRL
  rl from zero pretrain, can it be done? yes.
  ☆274Updated last week
• changjonathanc / flex-nano-vllm
  FlexAttention based, minimal vllm-style inference engine for fast Gemma 2 inference.
  ☆290Updated 2 months ago
• brendanhogan / DeepSeekRL-Extended
  Exploring Applications of GRPO
  ☆250Updated last month
• thinking-machines-lab / batch_invariant_ops
  ☆773Updated 3 weeks ago
• huggingface / picotron_tutorial
  ☆222Updated this week
• facebookresearch / PhysicsLM4
  Physics of Language Models, Part 4
  ☆247Updated 2 months ago
• wolfecameron / nanoMoE
  An extension of the nanoGPT repository for training small MOE models.
  ☆195Updated 6 months ago
• fangyuan-ksgk / Tiny-GRPO
  minimal GRPO implementation from scratch
  ☆98Updated 6 months ago
• microsoft / nanovppo
  Nano repo for RL training of LLMs
  ☆66Updated last month
• thinking-machines-lab / tinker-cookbook
  Post-training with Tinker
  ☆550Updated this week
• NVIDIA-NeMo / RL
  Scalable toolkit for efficient model reinforcement
  ☆910Updated this week
• microsoft / ArchScale
  Simple & Scalable Pretraining for Neural Architecture Research
  ☆296Updated last month
• MoonshotAI / checkpoint-engine
  Checkpoint-engine is a simple middleware to update model weights in LLM inference engines
  ☆751Updated last week
• JT-Ushio / MHA2MLA
  Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs
  ☆192Updated 3 months ago
• ypwang61 / One-Shot-RLVR
  [NeurIPS 2025] Reinforcement Learning for Reasoning in Large Language Models with One Training Example
  ☆360Updated last week
• facebookresearch / memory
  Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, spars…
  ☆343Updated 9 months ago
• PrimeIntellect-ai / prime-rl
  Async RL Training at Scale
  ☆669Updated this week
• ZihanWang314 / CoE
  Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models
  ☆220Updated 3 weeks ago
• ServiceNow / PipelineRL
  A scalable asynchronous reinforcement learning implementation with in-flight weight updates.
  ☆151Updated this week
• open-thought / tiny-grpo
  Minimal hackable GRPO implementation
  ☆286Updated 8 months ago
• weigao266 / Awesome-Efficient-Arch
  Speed Always Wins: A Survey on Efficient Architectures for Large Language Models
  ☆337Updated last month
• facebookresearch / RAM
  A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).
  ☆290Updated last week
• microsoft / dion
  Dion optimizer algorithm
  ☆360Updated last week
• groundlight / r1_vlm
  Build your own visual reasoning model
  ☆409Updated last month
• marin-community / marin
  Open-source framework for the research and development of foundation models.
  ☆466Updated this week
• s-sahoo / Eso-LMs
  Esoteric Language Models
  ☆99Updated 2 months ago
• joey00072 / nanoGRPO
  nanoGRPO is a lightweight implementation of Group Relative Policy Optimization (GRPO)
  ☆120Updated 4 months ago
• SakanaAI / RLT
  Training teachers with reinforcement learning able to make LLMs learn how to reason for test time scaling.
  ☆343Updated 3 months ago