AI-Hypercomputer / JetStream
JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs welcome).
☆229Updated this week
Related projects ⓘ
Alternatives and complementary repositories for JetStream
- ☆121Updated this week
- PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"☆40Updated this week
- Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimenta…☆457Updated 2 weeks ago
- Google TPU optimizations for transformers models☆74Updated this week
- ☆223Updated 4 months ago
- ☆155Updated this week
- xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerat…☆80Updated this week
- ☆178Updated last week
- Module, Model, and Tensor Serialization/Deserialization☆187Updated 3 weeks ago
- Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)☆64Updated 2 months ago
- ☆96Updated last month
- A high-throughput and memory-efficient inference and serving engine for LLMs☆250Updated last month
- ☆144Updated this week
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…☆146Updated last week
- JAX-Toolbox☆241Updated this week
- ☆262Updated this week
- CUDA checkpoint and restore utility☆220Updated 6 months ago
- Easy and lightning fast training of 🤗 Transformers on Habana Gaudi processor (HPU)☆152Updated this week
- Applied AI experiments and examples for PyTorch☆160Updated last week
- TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and sup…☆332Updated 3 weeks ago
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆163Updated this week
- seqax = sequence modeling + JAX☆132Updated 3 months ago
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆261Updated last year
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆479Updated 2 weeks ago
- Evaluate and Enhance Your LLM Deployments for Real-World Inference Needs☆160Updated last week
- ☆110Updated this week
- ☆64Updated 2 years ago
- A library to analyze PyTorch traces.☆300Updated this week
- A collection of YAML files, Helm Charts, Operator code, and guides to act as an example reference implementation for NVIDIA NIM deploymen…☆138Updated 2 weeks ago
- 🏋️ A unified multi-backend utility for benchmarking Transformers, Timm, PEFT, Diffusers and Sentence-Transformers with full support of O…☆251Updated last week