Lists (3)
Stars
Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals
nnScaler: Compiling DNN models for Parallel Training
Implementation of "Hyperspherical Latents Improve Continuous-Token Autoregressive Generation"
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…
Official inference framework for 1-bit LLMs
The leading native Python SSHv2 protocol library.
A PyTorch native platform for training generative AI models
New repo collection for NVIDIA Cosmos: https://github.com/nvidia-cosmos
A Data Streaming Library for Efficient Neural Network Training
Official inference library for pre-processing of Mistral models
Official inference library for Mistral models
[TMLR 2025🔥] A survey for the autoregressive models in vision.
🔥🔥🔥 [IEEE TCSVT] Latest Papers, Codes and Datasets on Vid-LLMs.
Intel® NPU Acceleration Library
A Python package for extending the official PyTorch that can easily obtain performance on Intel platform
ROCm / flash-attention
Forked from Dao-AILab/flash-attentionFast and memory-efficient exact attention
Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.
Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.
This repo contains the code for 1D tokenizer and generator
SEED-Voken: A Series of Powerful Visual Tokenizers
Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.