Ongoing research training transformer models at scale
☆15,242Feb 21, 2026Updated last week
Alternatives and similar repositories for Megatron-LM
Users that are interested in Megatron-LM are comparing it to the libraries listed below
Sorting:
- DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.☆41,648Updated this week
- Fast and memory-efficient exact attention☆22,361Updated this week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,229Aug 14, 2025Updated 6 months ago
- Transformer related optimization, including BERT, GPT☆6,394Mar 27, 2024Updated last year
- A high-throughput and memory-efficient inference and serving engine for LLMs☆71,234Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆23,658Updated this week
- verl: Volcano Engine Reinforcement Learning for LLMs☆19,339Updated this week
- Development repository for the Triton language and compiler☆18,460Feb 22, 2026Updated last week
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,170Feb 21, 2026Updated last week
- TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizat…☆12,938Updated this week
- Train transformer language models with reinforcement learning.☆17,460Updated this week
- A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Auto…☆16,807Updated this week
- 🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.☆20,678Updated this week
- An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & TIS & vLLM & Ray & Async RL)☆9,037Feb 21, 2026Updated last week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,436Mar 20, 2024Updated last year
- Example models using DeepSpeed☆6,785Feb 7, 2026Updated 3 weeks ago
- PyTorch extensions for high performance and large scale training.☆3,400Apr 26, 2025Updated 10 months ago
- 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (i…☆9,513Updated this week
- Hackable and optimized Transformers building blocks, supporting a composable construction.☆10,353Feb 20, 2026Updated last week
- Making large AI models cheaper, faster and more accessible☆41,359Updated this week
- Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities☆22,033Jan 23, 2026Updated last month
- A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch☆8,926Updated this week
- Facebook AI Research Sequence-to-Sequence Toolkit written in Python.☆32,170Sep 30, 2025Updated 5 months ago
- A framework for few-shot evaluation of language models.☆11,478Feb 15, 2026Updated 2 weeks ago
- 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal model…☆157,071Updated this week
- CUDA Templates and Python DSLs for High-Performance Linear Algebra☆9,315Updated this week
- Accessible large language models via k-bit quantization for PyTorch.☆7,997Updated this week
- FlashInfer: Kernel Library for LLM Serving☆5,009Updated this week
- A PyTorch native platform for training generative AI models☆5,098Updated this week
- An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.☆39,414Jun 2, 2025Updated 8 months ago
- Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)☆67,659Updated this week
- Large Language Model Text Generation Inference☆10,774Jan 8, 2026Updated last month
- Inference code for Llama models☆59,166Jan 26, 2025Updated last year
- The Triton Inference Server provides an optimized cloud and edge inferencing solution.☆10,393Updated this week
- Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.☆41,516Updated this week
- Optimized primitives for collective multi-GPU communication☆4,474Updated this week
- Code and documentation to train Stanford's Alpaca models, and generate the data.☆30,271Jul 17, 2024Updated last year
- An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries☆7,395Feb 3, 2026Updated 3 weeks ago
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆4,795Feb 22, 2026Updated last week