The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.
☆1,593Dec 15, 2025Updated 8 months ago
Alternatives and similar repositories for Pai-Megatron-Patch
Users that are interested in Pai-Megatron-Patch are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Best practice for training LLaMA models in Megatron-LM☆666Jan 2, 2024Updated 2 years ago
- Ongoing research training transformer models at scale☆17,774Updated this week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,261Aug 14, 2025Updated last year
- A flexible and efficient training framework for large-scale alignment tasks☆450Oct 23, 2025Updated 10 months ago
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆231Jun 15, 2026Updated 2 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,448Mar 20, 2024Updated 2 years ago
- Training library for Megatron-based models with bidirectional Hugging Face conversion capability☆904Updated this week
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆23,336Updated this week
- An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Asy…☆9,979Aug 13, 2026Updated 3 weeks ago
- Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.☆201May 29, 2026Updated 3 months ago
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,526Updated this week
- An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models☆3,387Updated this week
- VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo☆2,195Updated this week
- Ring attention implementation with flash attention☆1,053Sep 10, 2025Updated 11 months ago
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,358Aug 28, 2025Updated last year
- slime is an LLM post-training framework for RL Scaling.☆8,404Updated this week
- Zero Bubble Pipeline Parallelism☆466May 7, 2025Updated last year
- PyTorch bindings for CUTLASS grouped GEMM.☆192Apr 8, 2026Updated 5 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆691Updated this week
- Scalable toolkit for efficient model alignment☆852Oct 6, 2025Updated 11 months ago
- Minimalistic large language model 3D-parallelism training☆2,811May 26, 2026Updated 3 months ago
- Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL…☆15,525Updated this week
- The RL Bridge for LLM-based Agent Applications. Made Simple & Flexible.☆5,733Updated this week
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆1,037Mar 3, 2026Updated 6 months ago
- DLRover: An Automatic Distributed Deep Learning System☆1,682Updated this week
- distributed trainer for LLMs☆590May 20, 2024Updated 2 years ago
- OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, …☆7,397Updated this week
- ☆84Sep 9, 2023Updated 2 years ago
- Example models using DeepSpeed☆6,847Aug 26, 2026Updated last week
- A PyTorch native platform for training generative AI models☆5,708Updated this week
- Fast and memory-efficient exact attention☆24,856Updated this week
- Reproduce R1 Zero on Logic Puzzle☆2,450Mar 20, 2025Updated last year
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,516Updated this week
- Train transformer language models with reinforcement learning.☆19,240Updated this week
- Megatron's multi-modal data loader☆381Updated this week
- DeepEP: an efficient expert-parallel communication library☆10,106Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆35,594Updated this week
- EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL☆5,151Aug 31, 2026Updated last week
- BELLE: Be Everyone's Large Language model Engine（开源中文对话大模型）☆8,272Oct 16, 2024Updated last year