Alternatives and similar repositories for DeepSeek-R1-TrainingSuite

Users that are interested in DeepSeek-R1-TrainingSuite are comparing it to the libraries listed below

• 360CVGroup / 360VL
  Our 2nd-gen LMM
  ☆34Updated last year
• pangu-tech / pangu-ultra
  ☆74Updated 8 months ago
• StarRing2022 / R1-Nature
  最简易的R1结果在小模型上的复现，阐述类O1与DeepSeek R1最重要的本质。Think is all your need。利用实验佐证，对于强推理能力，think思考过程性内容是AGI/ASI的核心。
  ☆45Updated last year
• zhaochenyang20 / Prompt2Model-Self-Guide
  SELF-GUIDE: Better Task-Specific Instruction Following via Self-Synthetic Finetuning. COLM 2024 Accepted Paper
  ☆32Updated last year
• RUC-GSAI / YuLan-Mini
  A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.
  ☆223Updated 6 months ago
• bigai-nlco / TokenSwift
  [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation
  ☆121Updated 8 months ago
• SkyworkAI / Skywork-MoE
  Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models
  ☆139Updated last year
• GuoYiFantastic / IMelodist
  Music large model based on InternLM2-chat.
  ☆23Updated last year
• gabrielolympie / moe-pruner
  A repository aimed at pruning DeepSeek V3, R1 and R1-zero to a usable size
  ☆83Updated 5 months ago
• jackfsuia / nanoRLHF
  RLHF experiments on a single A100 40G GPU. Support PPO, GRPO, REINFORCE, RAFT, RLOO, ReMax, DeepSeek R1-Zero reproducing.
  ☆79Updated 11 months ago
• FreedomIntelligence / FastLLM
  Fast LLM Training CodeBase With dynamic strategy choosing [Deepspeed+Megatron+FlashAttention+CudaFusionKernel+Compiler];
  ☆40Updated 2 years ago
• BytedanceDouyinContent / SAIL-VL2
  The SAIL-VL2 series model developed by the BytedanceDouyinContent Group
  ☆76Updated 4 months ago
• kyegomez / OpenR1
  An open source implementation of R1
  ☆29Updated this week
• 18907305772 / FuseAI
  FuseAI Project
  ☆87Updated last year
• woct0rdho / transformers-qwen3-moe-fused
  Fused Qwen3 MoE layer for faster training, compatible with Transformers, LoRA, bnb 4-bit quant, Unsloth. Also possible to train LoRA over…
  ☆231Updated this week
• zai-org / GLM-Edge
  GLM Series Edge Models
  ☆158Updated 7 months ago
• OpenGVLab / Docopilot
  [CVPR 2025] Docopilot: Improving Multimodal Models for Document-Level Understanding
  ☆36Updated 6 months ago
• xverse-ai / XVERSE-MoE-A36B
  XVERSE-MoE-A36B: A multilingual large language model developed by XVERSE Technology Inc.
  ☆38Updated last year
• QwenLM / WorldPM
  ☆93Updated 8 months ago
• Bui1dMySea / MemLong
  ☆96Updated last year
• yyht / openrlhf_async_pipline
  ☆87Updated 5 months ago
• wajihullahbaig / deepseekv3-minimal
  Creating the DeepSeek V3 model from scratch
  ☆24Updated 10 months ago
• Emericen / tiny-qwen
  A minimal PyTorch re-implementation of Qwen3 VL with a fancy CLI
  ☆316Updated 2 months ago
• saxenarohit / MovieSum
  ☆19Updated last year
• thunlp / Delta-CoMe
  Delta-CoMe can achieve near loss-less 1-bit compressin which has been accepted by NeurIPS 2024
  ☆58Updated last year
• WALLE-AI / uReasoningLLMs
  Deepseek-r1复现科普与资源汇总
  ☆22Updated 11 months ago
• inclusionAI / AWorld-RL
  Agentic Learning Powered by AWorld
  ☆86Updated last week
• InternLM / Condor
  [ACL 2025] An official pytorch implement of the paper: Condor: Enhance LLM Alignment with Knowledge-Driven Data Synthesis and Refinement
  ☆40Updated 8 months ago
• ArtificialZeng / llama3_explained
  the newest version of llama3，source code explained line by line using Chinese
  ☆22Updated last year
• Lauorie / DFT
  Reproduced the DFT method without using Verl. https://arxiv.org/abs/2508.05629
  ☆21Updated 3 months ago