Search Results for "sparse"
Sort By:
26 projects for "sparse" with 2 filters applied:
-
Ship Agents FasterGemini Enterprise Agent Platform lets you rapidly build, scale, govern and optimize production-ready agents grounded in your organization's data. The platform enables developers to build custom or pre-built agents for virtually any use case. New customers get $300 in free credits.
-
Enterprise-grade ITSM, for every businessFreshservice is an intuitive, AI-powered platform that helps IT, operations, and business teams deliver exceptional service without the usual complexity. Automate repetitive tasks, resolve issues faster, and provide seamless support across the organization. From managing incidents and assets to driving smarter decisions, Freshservice makes it easy to stay efficient and scale with confidence.
-
1
DeepSeek-V3.2-Exp
An experimental version of DeepSeek model
DeepSeek-V3.2-Exp is an experimental release of the DeepSeek model family, intended as a stepping stone toward the next generation architecture. The key innovation in this version is DeepSeek Sparse Attention (DSA), a sparse attention mechanism that aims to optimize training and inference efficiency in long-context settings without degrading output quality. According to the authors, they aligned the training setup of V3.2-Exp with V3.1-Terminus so that benchmark results remain largely comparable, even though the internal attention mechanism changes. ... -
2
PySINDy
A package for the sparse identification of nonlinear dynamical systems
PySINDy is a Python library that implements the Sparse Identification of Nonlinear Dynamics (SINDy) method for discovering mathematical models of dynamical systems from data. The framework focuses on identifying governing equations that describe the behavior of complex physical systems by selecting sparse combinations of candidate functions. Instead of fitting a purely predictive machine learning model, PySINDy attempts to recover interpretable differential equations that explain how a system evolves over time. ... -
3
DLRM
An implementation of a deep learning recommendation model (DLRM)
DLRM (Deep Learning Recommendation Model) is Meta’s open-source reference implementation for large-scale recommendation systems built to handle extremely high-dimensional sparse features and embedding tables. The architecture combines dense (MLP) and sparse (embedding) branches, then interacts features via dot product or feature interactions before passing through further dense layers to predict click-through, ranking scores, or conversion probabilities. The implementation is optimized for performance at scale, supporting multi-GPU and multi-node execution, quantization, embedding partitioning, and pipelined I/O to feed huge embeddings efficiently. ... -
4
GLM-5
From Vibe Coding to Agentic Engineering
...Building on earlier GLM series models, GLM-5 dramatically scales the parameter count (to roughly 744 billion) and expands pre-training data to significantly improve performance on complex tasks such as multi-step reasoning, software engineering workflows, and agent orchestration compared to its predecessors like GLM-4.5. It incorporates innovations like DeepSeek Sparse Attention (DSA) to preserve massive context windows while reducing deployment costs and supporting long context processing, which is crucial for detailed plans and agent tasks. -
Secure File Transfer for Windows with Cerberus by RedwoodCerberus supports unlimited users and connections on a single IP, with built-in encryption, 2FA, and a browser-based web client — all deployable in under 15 minutes with a 25-day free trial.
-
5
MiniCPM4.1
Achieving 3+ generation speedup on reasoning tasks
...One of its key innovations is the hybrid reasoning mode, which allows developers to control whether the model engages in deeper reasoning processes or faster responses depending on the use case. The model also supports both dense and sparse attention mechanisms, enabling more efficient computation depending on the selected inference framework. With improved pretraining on longer sequences and enhanced scaling techniques, MiniCPM4.1 delivers better performance in long-context tasks and complex problem solving. -
6
Ling-V2
Ling-V2 is a MoE LLM provided and open-sourced by InclusionAI
Ling-V2 is an open-source family of Mixture-of-Experts (MoE) large language models developed by the InclusionAI research organization with the goal of combining state-of-the-art performance, efficiency, and openness for next-generation AI applications. It introduces highly sparse architectures where only a fraction of the model’s parameters are activated per input token, enabling models like Ling-mini-2.0 to achieve reasoning and instruction-following capabilities on par with much larger dense models while remaining significantly more computationally efficient. Trained on more than 20 trillion tokens of high-quality data and enhanced through multi-stage supervised fine-tuning and reinforcement learning, Ling-V2’s models demonstrate strong general reasoning, mathematical problem-solving, coding understanding, and knowledge-intensive task performance. -
7
TurboDiffusion
100–200× Acceleration for Video Diffusion Models
TurboDiffusion is an advanced open-source framework designed to dramatically accelerate video diffusion model generation, aiming for performance improvements on the order of 100–200× compared with traditional implementations while retaining high output quality. It achieves this by combining a suite of algorithmic and engineering optimizations, including attention acceleration techniques, efficient step distillation methods, and quantization strategies that reduce computational overhead. The... -
8
HunyuanImage-3.0
A Powerful Native Multimodal Model for Image Generation
HunyuanImage-3.0 is a powerful, native multimodal text-to-image generation model released by Tencent’s Hunyuan team. It unifies multimodal understanding and generation in a single autoregressive framework, combining text and image modalities seamlessly rather than relying on separate image-only diffusion components. It uses a Mixture-of-Experts (MoE) architecture with many expert subnetworks to scale efficiently, deploying only a subset of experts per token, which allows large parameter... -
9
3FS
A high-performance distributed file system
...Its primary aim is to support efficient and scalable feature transformation pipelines—especially for inference environments—by batching, caching, and integrating feature-based modules like segmenters, sparse retrievers, and scorers seamlessly. The repo includes APIs to define components (e.g. seg, ret, scor) that wrap or interface with external or internal modules, as well as logic to schedule and compose these feature transforms. By handling caching and batching at a system level, 3FS helps reduce overhead when many features or modules must be evaluated per input (e.g. in an LLM agent pipeline). ... -
Our Free Plans just got better! | Auth0You asked, we delivered! Auth0 is excited to expand our Free and Paid plans to include more options so you can focus on building, deploying, and scaling applications without having to worry about your security. Auth0 now, thank yourself later.
-
10
kg-gen
Knowledge Graph Generation from Any Text
kg-gen is an open-source framework developed by the STAIR Lab that automatically generates knowledge graphs from unstructured text using large language models. The system is designed to transform plain text sources such as documents, articles, or conversation transcripts into structured graphs composed of entities and relationships. Instead of relying on traditional rule-based extraction techniques, KG-Gen uses language models to identify entities and their relationships, producing... -
11
Map-Anything
MapAnything: Universal Feed-Forward Metric 3D Reconstruction
...Instead of stitching together many task-specific models, it uses a single architecture that supports a wide range of 3D tasks—multi-image structure-from-motion, multi-view stereo, monocular metric depth, registration, depth completion, and more. The model flexibly accepts different input combinations (images, intrinsics, poses, sparse or dense depth) and produces a rich set of outputs including per-pixel 3D points, camera intrinsics, camera poses, ray directions, confidence maps, and validity masks. Its inference path is fully feed-forward with optional mixed-precision and memory-efficient modes, making it practical to scale to long image sequences while keeping latency predictable. -
12
DeepSpeed
Deep learning optimization library: makes distributed training easy
DeepSpeed is an easy-to-use deep learning optimization software suite that enables unprecedented scale and speed for Deep Learning Training and Inference. With DeepSpeed you can: 1. Train/Inference dense or sparse models with billions or trillions of parameters 2. Achieve excellent system throughput and efficiently scale to thousands of GPUs 3. Train/Inference on resource constrained GPU systems 4. Achieve unprecedented low latency and high throughput for inference 5. Achieve extreme compression for an unparalleled inference latency and model size reduction with low costs DeepSpeed offers a confluence of system innovations, that has made large scale DL training effective, and efficient, greatly improved ease of use, and redefined the DL training landscape in terms of scale that is possible. ... -
13
Armadillo
fast C++ library for linear algebra & scientific computing
* Fast C++ library for linear algebra (matrix maths) and scientific computing * Easy to use functions and syntax, deliberately similar to Matlab / Octave * Uses template meta-programming techniques to increase efficiency * Provides user-friendly wrappers for OpenBLAS, Intel MKL, LAPACK, ATLAS, ARPACK, SuperLU and FFTW libraries * Useful for machine learning, pattern recognition, signal processing, bioinformatics, statistics, finance, etc. * Downloads:... -
14
Mixtral offloading
Run Mixtral-8x7B models in Colab or consumer desktops
...The project implements techniques that allow model components to be dynamically moved between CPU memory and GPU memory during inference, significantly reducing the amount of GPU VRAM required to run the model. This approach takes advantage of the sparse activation properties of mixture-of-experts architectures, where only a subset of expert networks are used for each token during generation. By selectively loading and caching the required experts, the system avoids keeping the entire model in GPU memory at once. The repository includes notebooks and code examples that demonstrate how to run large language models on consumer hardware such as personal GPUs or cloud notebook environments. -
15
LLaMA-MoE
Building Mixture-of-Experts from LLaMA with Continual Pre-training
...The repository is centered on making MoE research more accessible by offering smaller and more affordable models with only about 3.0 to 3.5 billion activated parameters, which helps reduce deployment and experimentation costs. Its architecture works by splitting LLaMA feed-forward networks into sparse experts and adding gating mechanisms so that only selected experts are activated during inference and training. The project is not just a model release, but also a research framework that includes multiple expert construction methods, several gating strategies, and tooling for continual pre-training on filtered SlimPajama-based datasets. ... -
16
CoTracker
CoTracker is a model for tracking any point (pixel) on a video
...By reasoning about all tracks together, it can maintain temporal consistency, handle mutual occlusions, and reduce identity swaps when trajectories cross. The model takes sparse point queries on one frame and predicts their sub-pixel locations and a visibility score for every subsequent frame, producing long, coherent trajectories. Its transformer-style architecture aggregates information both along time and across points, allowing it to recover tracks even after brief disappearances. The repository ships with inference scripts, pretrained weights, and simple interfaces to seed points, run tracking, and export trajectories for downstream tasks. ... -
17
ekho
Chinese text-to-speech engine
ekho is a project with relatively sparse documentation, but from the repository it appears to be a small-scale tool for audio processing and playback, possibly with features for speech synthesis or manipulation. The repo includes scripts and configuration files suggesting interactions with media/audio handling libraries. Because of limited README detail, it seems targeted at users comfortable reading and modifying code, rather than end users expecting polished UIs. -
18
MLPACK is a C++ machine learning library with emphasis on scalability, speed, and ease-of-use. Its aim is to make machine learning possible for novice users by means of a simple, consistent API, while simultaneously exploiting C++ language features to provide maximum performance and flexibility for expert users. * More info + downloads: https://mlpack.org * Git repo: https://github.com/mlpack/mlpack
-
19
Immutable Sparse Wave Trees (WaveTree)
Realtime bigdata tool for bit strings up to 2^63 based on AVL forest
...All those operations can be done millions of times per second regardless of size because the AVL forest reuses existing branches recursively. Theres a scalar (originally for copy/pasting subranges of sounds) and a bit Java package. Sparse n dimensional matrix. -
20
This project provides a set of Python tools for creating various kinds of neural networks, which can also be powered by genetic algorithms using grammatical evolution. MLP, backpropagation, recurrent, sparse, and skip-layer networks are supported.
-
21
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.
-
22
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.
-
23
Command A+
4-bit Command A+ model for enterprise agents and multilingual tasks
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. ... -
24
DeepSeek-V3.2-Speciale
High-compute ultra-reasoning model surpassing model surpassing GPT-5
DeepSeek-V3.2-Speciale is the high-compute, ultra-reasoning variant of DeepSeek-V3.2, designed specifically to push the boundaries of mathematical, logical, and algorithmic intelligence. It builds on the DeepSeek Sparse Attention (DSA) framework, delivering dramatically improved long-context efficiency while preserving full model quality. Unlike the standard version, Speciale is tuned exclusively for deep reasoning and therefore does not support tool-calling, focusing its full capacity on pure cognitive performance. The model uses a scaled reinforcement learning framework that allows it to surpass GPT-5 in several evaluations and reach reasoning performance comparable to Gemini-3.0-Pro. ... -
25
DeepSeek-V3.2
High-efficiency reasoning and agentic intelligence model
DeepSeek-V3.2 is a cutting-edge large language model developed by DeepSeek-AI, focused on achieving high reasoning accuracy and computational efficiency for agentic tasks. It introduces DeepSeek Sparse Attention (DSA), a new attention mechanism that dramatically reduces computational overhead while maintaining strong long-context performance. Built with a scalable reinforcement learning framework, it reaches near-GPT-5 levels of reasoning and outperforms comparable models like DeepSeek-V3.1 and Gemini-3.0-Pro in advanced benchmarks. ...
