Search Results for "sparse" - Page 7

The Lidy project is a small cross-platform library under LPGL licence, written in C language. It has been designed in order to facilitate the sparse matrices handling.

The JSparse Matrix Package, developed by Philipp Geigenmüller during an internship at the prudsys AG in Chemnitz, Germany, is an extension of the well-known Java Matrix Package (JAMA) and allows the use of sparse matrices and related algorithms.

PySparse - A Sparse Matrix Library for Python

A simple CUDA accelerated solver for general dimensional n-body problems. Currently only the conjugate gradient method has been implemented, but more will soon follow.

I/O routines and tools for sparse matrices.

LSGTL means LLX’s Static Graph Template Library which is a light-weighted header-only template library developed mainly for static graph analysis. LSGTL is expected to be used in laboratories for research purposes mostly.

develop a computer algebra system that does sparse polynomial and matrices on cbe

Pydusa is a package for parallel programming using Python. It contains a module for doing MPI programming in Python. We have added parallel solver packages such as Parallel SuperLU for solving sparse linear systems.

CSparseJ is a Java port of CSparse: a Concise Sparse matrix package.

vmdkutil is a utility to manage VMware sparse virtual disk files (.vmdk), creating, cloning, merging, etc.

The Optimized Sparse Kernel Interface (OSKI) Library provides automatically tuned sparse matrix kernels, for use by solver libraries and applications. OSKI is part of the BeBOP project on performance tuning and analysis at U.C. Berkeley. (Go Bears!)

Sparse lets you write simple HTML templates which are turned into fully-fledged MySQL-based applications with degradable Ajax. Easily integrated with PHP code. Does automated data and error handling, input validation, searching, sorting, and navigation.

BPKIT is a toolkit of block preconditioners for the iterative solution of sparse linear systems.

JMatrixView is aJava/Swing matrix viewer. It works with Colt matrix implementation. It features dynamically tunable row/column resolution and has been designed for sparse matrix.

...It provides the possibility to add support for any compiler! For the latest news and versions please visit the alt-ide website at www.alt-ide.net as updates on SF.net are sparse!

An ANN design/implementation focused to work with optimally with NEAT (http://www.cs.utexas.edu/users/kstanley/neat.html). Semiann uses a sparse matrix to fill out the nodes, connections, weights, biases and activation functions for an ANN.

uBLAS is a C++ template class library that provides BLAS level 1, 2, 3 functionality for dense, packed and sparse matrices and many numerical algorithms for linear algebra.

The goal of this project is to produce a genuinely Open Source multicast router that supports all existing multicast routing protocols. This will include DVMRP, PIM-2 (sparse and dense modes), MOSPF and BGMP.

sparse linear algebra library, with emphasis on preconditioned Krylov subspace methods for the solution of linear systems of equations

MatGen is a matrix generator. It outputs matrices in MatrixMarket file format, taking as input a file describing the matrices´ properties. Currently it supports only random sparse matrices.

Kimi K3 is an open-weight, multim is an open-weight, multimodal agentic AI model from Moonshot AI designed for advanced coding, research, reasoning, and knowledge work. Builtodal agentic AI model from Moonshot AI designed for advanced coding, research, reasoning, and knowledge work. Built with 2.8 trillion total parameters and a sparse mixture-of-experts architecture, it activates 104 billion parameters per token to with 2.8 trillion total parameters and a sparse mixture-of-experts architecture, it activates 104 billion parameters per token to deliver frontier-level performance more efficiently. Kimi K3 combines native text and image understanding deliver frontier-level performance more efficiently.

Blit contains a group of highly efficient iterative sparse solvers that can handle multiple right-hand-sides (i.e. block solvers). We will implement BLQMR, BLGMRES and other block algorithms in MATLAB, FORTRAN 90, C/C++, CUDA and OpenCL.

LongCat-2.0 is Meituan’s flagship open-weight Mixture-of-Experts language model designed for frontier-scale coding, reasoning, and autonomous agent workflows. It features 1.6 trillion total parameters with approximately 48 billion activated per token, combining high capability with efficient sparse inference. The model was pretrained on more than 35 trillion tokens and trained entirely on a large-scale cluster of domestically developed AI accelerators, demonstrating stable frontier-scale training without rollback events. LongCat-2.0 introduces LongCat Sparse Attention and extensive 1M-context training, enabling native processing of million-token inputs for long-document analysis, repository-scale coding, and complex multi-step reasoning. ...

Inkling-Small is an open-weight general-purpose multimodal model from Thinking Machines Lab, designed for agentic systems, coding assistants, chatbots, retrieval workflows, and natural-language applications. It accepts text, images, and audio as input and produces text output, with multilingual and multi-programming-language capabilities. The model uses a sparse Mixture-of-Experts architecture with 276B total parameters and 12B active per token, enabling strong performance with lower inference cost than a fully dense model of similar scale. Its 42-layer decoder routes each token through six of 256 specialized experts plus two shared experts, while hybrid local and global attention supports efficient processing. ...

Command A+ 05-2026 W4A4 is a 4-bit quantized version of Cohere’s open-source Command A+ model, optimized for enterprise-grade agentic, multilingual, and reasoning-heavy workloads. It supports text and image inputs, generates text outputs, and uses a sparse Mixture-of-Experts Transformer architecture with 218B total parameters and 25B active parameters. The W4A4 release applies 4-bit weight and activation quantization mainly to MoE experts, preserving attention components at full precision to reduce quality loss while improving speed, latency, and hardware efficiency. Cohere recommends W4A4 for most users because it offers a smaller hardware footprint with negligible benchmark differences compared to BF16 and FP8 versions. ...