huggingface / Megatron-LMLinks
Ongoing research training transformer models at scale
☆18Updated 2 years ago
Alternatives and similar repositories for Megatron-LM
Users that are interested in Megatron-LM are comparing it to the libraries listed below
Sorting:
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆70Updated 2 years ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆112Updated 9 months ago
- ☆79Updated 2 years ago
- Train llm (bloom, llama, baichuan2-7b, chatglm3-6b) with deepspeed pipeline mode. Faster than zero/zero++/fsdp.☆98Updated last year
- Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models☆137Updated last year
- 📑 Dive into Big Model Training☆116Updated 3 years ago
- [ICML'24] The official implementation of “Rethinking Optimization and Architecture for Tiny Language Models”☆126Updated 11 months ago
- Fast LLM Training CodeBase With dynamic strategy choosing [Deepspeed+Megatron+FlashAttention+CudaFusionKernel+Compiler];☆41Updated last year
- Implementation of NAACL 2024 Outstanding Paper "LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models"☆152Updated 9 months ago
- Low-bit optimizers for PyTorch☆134Updated 2 years ago
- Scalable PaLM implementation of PyTorch☆189Updated 3 years ago
- Teacher - student distillation using DeepSpeed☆19Updated 3 years ago
- code for Scaling Laws of RoPE-based Extrapolation☆73Updated 2 years ago
- Linear Attention Sequence Parallelism (LASP)☆88Updated last year
- Model Compression for Big Models☆165Updated 2 years ago
- train llama on a single A100 80G node using 🤗 transformers and 🚀 Deepspeed Pipeline Parallelism☆225Updated 2 years ago
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆69Updated 2 years ago
- ☆108Updated 5 months ago
- Efficient, Flexible, and Highly Fault-Tolerant Model Service Management Based on SGLang☆61Updated last year
- Odysseus: Playground of LLM Sequence Parallelism☆78Updated last year
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆135Updated last year
- Official implementation of TransNormerLLM: A Faster and Better LLM☆248Updated last year
- A prototype repo for hybrid training of pipeline parallel and distributed data parallel with comments on core code snippets. Feel free to…☆57Updated 2 years ago
- A (somewhat) minimal library for finetuning language models with PPO on human feedback.☆88Updated 3 years ago
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆219Updated last year
- Repository of LV-Eval Benchmark☆72Updated last year
- A LLaMA1/LLaMA12 Megatron implement.☆28Updated 2 years ago
- REST: Retrieval-Based Speculative Decoding, NAACL 2024☆212Updated 3 months ago
- ☆157Updated 2 years ago
- This is a personal reimplementation of Google's Infini-transformer, utilizing a small 2b model. The project includes both model and train…☆58Updated last year