Oneflow-Inc / libaiLinks
LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training
☆405Updated 6 months ago
Alternatives and similar repositories for libai
Users that are interested in libai are comparing it to the libraries listed below
Sorting:
- Models and examples built with OneFlow☆101Updated last year
- ☆219Updated 2 years ago
- Best practice for training LLaMA models in Megatron-LM☆664Updated 2 years ago
- ☆79Updated 2 years ago
- Efficient Training (including pre-training and fine-tuning) for Big Models☆618Updated 3 months ago
- optimized BERT transformer inference on NVIDIA GPU. https://arxiv.org/abs/2210.03052☆476Updated last year
- Transformer related optimization, including BERT, GPT☆59Updated 2 years ago
- Easy Parallel Library (EPL) is a general and efficient deep learning framework for distributed model training.☆271Updated 2 years ago
- ☆130Updated last year
- OneFlow models for benchmarking.☆104Updated last year
- Transformer related optimization, including BERT, GPT☆39Updated 2 years ago
- Running BERT without Padding☆478Updated 3 years ago
- Collaborative Training of Large Language Models in an Efficient Way☆417Updated last year
- DeepLearning Framework Performance Profiling Toolkit☆296Updated 3 years ago
- Efficient Inference for Big Models☆585Updated 3 years ago
- oneflow documentation☆69Updated last year
- Model Compression for Big Models☆167Updated 2 years ago
- FlagScale is a large model toolkit based on open-sourced projects.☆471Updated this week
- Simple Dynamic Batching Inference☆145Updated 3 years ago
- ☆624Updated last month
- A flexible and efficient training framework for large-scale alignment tasks☆447Updated 3 months ago
- Examples of training models with hybrid parallelism using ColossalAI☆339Updated 2 years ago
- ☆74Updated last week
- ☆523Updated last week
- Large-scale model inference.☆627Updated 2 years ago
- InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencie…☆417Updated 5 months ago
- Accelerate inference without tears☆372Updated last week
- Train llm (bloom, llama, baichuan2-7b, chatglm3-6b) with deepspeed pipeline mode. Faster than zero/zero++/fsdp.☆98Updated 2 years ago
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆71Updated 2 years ago
- The road to hack SysML and become an system expert☆510Updated last year