(best/better) practices of megatron on veRL and tuning guide
☆141May 12, 2026Updated 6 months ago
Alternatives and similar repositories for verl_megatron_practice
Users that are interested in verl_megatron_practice are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆234Oct 31, 2026Updated last week
- verl: Volcano Engine Reinforcement Learning for LLMs☆14Nov 3, 2026Updated last week
- Ongoing research training transformer models at scale☆13Nov 5, 2026Updated last week
- Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.☆206May 29, 2026Updated 5 months ago
- Async pipelined version of Verl☆123Apr 8, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [Archived] For the latest updates and community contribution, please visit: https://github.com/Ascend/TransferQueue or https://gitcode.co…☆15Aug 14, 2026Updated 2 months ago
- ☆12Nov 21, 2025Updated 11 months ago
- Training library for Megatron-based models with bidirectional Hugging Face conversion capability☆991Updated this week
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆3,301Updated this week
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆697Sep 16, 2026Updated last month
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆123Dec 17, 2025Updated 10 months ago
- ☆292Updated this week
- Ring attention implementation with flash attention☆1,068Sep 10, 2025Updated last year
- ☆91Aug 16, 2025Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale☆676Updated this week
- Implementation for FP8/INT8 Rollout for RL training without performence drop.☆309Nov 7, 2025Updated last year
- Utility scripts for PyTorch (e.g. Make Perfetto show some disappearing kernels, Memory profiler that understands more low-level allocatio…☆115Sep 11, 2025Updated last year
- An LLM post-training framework with vLLM for RL Scaling☆504Updated this week
- ☆14May 9, 2024Updated 2 years ago
- ☆37Aug 7, 2025Updated last year
- 🔥 LLM-powered GPU kernel synthesis: Train models to convert PyTorch ops into optimized Triton kernels via SFT+RL. Multi-turn compilation…☆154Nov 10, 2025Updated last year
- [ICLR 2026] End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning☆408Mar 30, 2026Updated 7 months ago
- 🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support☆1,082Updated this week
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- slime is an LLM training framework for RL Scaling.☆8,818Nov 5, 2026Updated last week
- Allow torch tensor memory to be released and resumed later☆283Updated this week
- A flexible and efficient training framework for large-scale alignment tasks☆451Oct 23, 2025Updated last year
- Compact and Agent-Native MoE Training System☆360Nov 6, 2026Updated last week
- Checkpoint-engine is a simple middleware to update model weights in LLM inference engines☆1,023Oct 22, 2026Updated 3 weeks ago
- ☆13Oct 22, 2025Updated last year
- Sequence-level 1F1B schedule for LLMs.☆37Aug 26, 2025Updated last year
- An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models☆3,455Updated this week
- My learning notes for ML SYS.☆7,665Oct 21, 2026Updated 3 weeks 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.
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,281Updated this week
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆52Jul 4, 2025Updated last year
- An asynchronous streaming data management module for efficient post-training.☆175Updated this week
- The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.☆1,602Dec 15, 2025Updated 10 months ago
- Unleashing the Power of Reinforcement Learning for Math and Code Reasoners☆738Jun 6, 2025Updated last year
- Toolchain built around the Megatron-LM for Distributed Training☆100May 20, 2026Updated 5 months ago
- A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training☆973Updated this week
Are these results useful?