deepspeedai / Megatron-DeepSpeedLinks
Ongoing research training transformer language models at scale, including: BERT & GPT-2
☆2,193Updated 3 months ago
Alternatives and similar repositories for Megatron-DeepSpeed
Users that are interested in Megatron-DeepSpeed are comparing it to the libraries listed below
Sorting:
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,427Updated last year
- The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.☆1,436Updated last week
- Best practice for training LLaMA models in Megatron-LM☆660Updated last year
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,077Updated 4 months ago
- Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".☆2,219Updated last year
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,302Updated 8 months ago
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,347Updated 4 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…☆2,925Updated this week
- Efficient Training (including pre-training and fine-tuning) for Big Models☆612Updated 3 weeks ago
- LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalabili…☆3,730Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,559Updated last year
- Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 using FP8/NVFP4/MXFP4☆940Updated last week
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,664Updated last year
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,840Updated this week
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,272Updated 6 months ago
- S-LoRA: Serving Thousands of Concurrent LoRA Adapters☆1,868Updated last year
- A fast MoE impl for PyTorch☆1,813Updated 9 months ago
- Fast inference from large lauguage models via speculative decoding☆856Updated last year
- Transformer related optimization, including BERT, GPT☆6,354Updated last year
- Scalable toolkit for efficient model alignment☆845Updated last month
- ⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)☆994Updated 11 months ago
- A PyTorch Native LLM Training Framework☆884Updated 2 months ago
- LongBench v2 and LongBench (ACL 25'&24')☆1,020Updated 10 months ago
- A family of open-sourced Mixture-of-Experts (MoE) Large Language Models☆1,635Updated last year
- Open Academic Research on Improving LLaMA to SOTA LLM☆1,614Updated 2 years ago
- FlashInfer: Kernel Library for LLM Serving☆4,099Updated this week
- Ring attention implementation with flash attention☆910Updated 2 months ago
- PyTorch extensions for high performance and large scale training.☆3,387Updated 6 months ago
- Central place for the engineering/scaling WG: documentation, SLURM scripts and logs, compute environment and data.☆1,007Updated last year
- YaRN: Efficient Context Window Extension of Large Language Models☆1,637Updated last year