google / paxmlLinks
Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.
☆539Updated 2 months ago
Alternatives and similar repositories for paxml
Users that are interested in paxml are comparing it to the libraries listed below
Sorting:
- ☆190Updated 2 weeks ago
- jax-triton contains integrations between JAX and OpenAI Triton☆432Updated 3 weeks ago
- ☆337Updated this week
- ☆145Updated this week
- Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax☆680Updated this week
- JAX-Toolbox☆359Updated this week
- Orbax provides common checkpointing and persistence utilities for JAX users☆443Updated this week
- ☆545Updated last year
- ☆363Updated last year
- JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs wel…☆387Updated 4 months ago
- Library for reading and processing ML training data.☆586Updated this week
- seqax = sequence modeling + JAX☆168Updated 3 months ago
- MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvement…☆402Updated 2 weeks ago
- JAX Synergistic Memory Inspector☆179Updated last year
- ☆359Updated last week
- Implementation of Flash Attention in Jax☆220Updated last year
- ☆283Updated last year
- JAX implementation of the Llama 2 model☆215Updated last year
- Inference code for LLaMA models in JAX☆119Updated last year
- Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)☆441Updated 2 weeks ago
- Accelerate, Optimize performance with streamlined training and serving options with JAX.☆320Updated this week
- CLU lets you write beautiful training loops in JAX.☆358Updated 4 months ago
- TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and sup…☆399Updated this week
- ☆271Updated this week
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆580Updated 2 months ago
- JMP is a Mixed Precision library for JAX.☆208Updated 9 months ago
- Minimal yet performant LLM examples in pure JAX☆193Updated last month
- Train very large language models in Jax.☆209Updated 2 years ago
- For optimization algorithm research and development.☆542Updated last week
- A library to analyze PyTorch traces.☆421Updated last week