AI-Hypercomputer / jetstream-pytorch
PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"
☆59Updated 2 weeks ago
Alternatives and similar repositories for jetstream-pytorch:
Users that are interested in jetstream-pytorch are comparing it to the libraries listed below
- Google TPU optimizations for transformers models☆107Updated 2 months ago
- JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs wel…☆313Updated this week
- ☆137Updated this week
- Applied AI experiments and examples for PyTorch☆256Updated 3 weeks ago
- Fast low-bit matmul kernels in Triton☆285Updated this week
- extensible collectives library in triton☆84Updated last week
- KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA problems☆263Updated this week
- Write a fast kernel and run it on Discord. See how you compare against the best!☆38Updated this week
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆192Updated this week
- This repository contains the experimental PyTorch native float8 training UX☆222Updated 8 months ago
- ☆198Updated this week
- ring-attention experiments☆129Updated 5 months ago
- ☆103Updated 7 months ago
- ☆186Updated this week
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆238Updated this week
- ☆205Updated 2 months ago
- Cataloging released Triton kernels.☆213Updated 3 months ago
- ☆66Updated 2 weeks ago
- Load compute kernels from the Hub☆113Updated this week
- Collection of kernels written in Triton language☆118Updated last week
- QUICK: Quantization-aware Interleaving and Conflict-free Kernel for efficient LLM inference☆116Updated last year
- KV cache compression for high-throughput LLM inference☆125Updated 2 months ago
- Perplexity GPU Kernels☆185Updated last week
- ☆295Updated this week
- ☆97Updated this week
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆108Updated this week
- Experiment of using Tangent to autodiff triton☆78Updated last year
- PyTorch per step fault tolerance (actively under development)☆274Updated this week
- A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS☆156Updated 3 weeks ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆125Updated 4 months ago