Stars
Extract structured strategy specifications from quantitative finance research papers — Agent Skill for GitHub Copilot & Claude Code
upload big files to Zenodo using cURL, jq and bash
Fast OS-level support for GPU checkpoint and restore
ByteCheckpoint: An Unified Checkpointing Library for LFMs
Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
An adaptive collective communication library for distributed training
Triton-based implementation of Sparse Mixture of Experts.
Official Repo for "SplitQuant / LLM-PQ: Resource-Efficient LLM Offline Serving on Heterogeneous GPUs via Phase-Aware Model Partition and Adaptive Quantization"
[ICML 2024] CLLMs: Consistency Large Language Models
Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines
[CVPR 2024] Prompt Highlighter: Interactive Control for Multi-Modal LLMs
[ECCV 2024, Oral] DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors
Distributed Deep Graph Learning Framework for Dynamic Graphs
S-LoRA: Serving Thousands of Concurrent LoRA Adapters
[NeurIPS 2023 Spotlight] Real-World Image Variation by Aligning Diffusion Inversion Chain
TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. Tensor…
A resilient distributed training framework
BladeDISC is an end-to-end DynamIc Shape Compiler project for machine learning workloads.
Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads
Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals
A Survey on multimodal learning research.
mPLUG-Owl: The Powerful Multi-modal Large Language Model Family
Adaptive Message Quantization and Parallelization for Distributed Full-graph GNN Training
An open-source framework for training large multimodal models.