Search Results for "mixture" - Page 2
Sort By:
Showing 55 open source projects for "mixture"
View related business solutions-
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.
-
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
1
Tencent-Hunyuan-Large
Open-source large language model family from Tencent Hunyuan
Tencent-Hunyuan-Large is the flagship open-source large language model family from Tencent Hunyuan, offering both pre-trained and instruct (fine-tuned) variants. It is designed with long-context capabilities, quantization support, and high performance on benchmarks across general reasoning, mathematics, language understanding, and Chinese / multilingual tasks. It aims to provide competitive capability with efficient deployment and inference. FP8 quantization support to reduce memory usage... -
2
Hivemind
Decentralized deep learning in PyTorch. Built to train models
...Decentralized parameter averaging: iteratively aggregate updates from multiple workers without the need to synchronize across the entire network. Train neural networks of arbitrary size: parts of their layers are distributed across the participants with the Decentralized Mixture-of-Experts. If you have succesfully trained a model or created a downstream repository with the help of our library, feel free to submit a pull request that adds your project to the list. -
3
DeepSeek VL2
Mixture-of-Experts Vision-Language Models for Advanced Multimodal
DeepSeek-VL2 is DeepSeek’s vision + language multimodal model—essentially the next-gen successor to their first vision-language models. It combines image and text inputs into a unified embedding / reasoning space so that you can query with text and image jointly (e.g. “What’s going on in this scene?” or “Generate a caption appropriate to context”). The model supports both image understanding (vision tasks) and multimodal reasoning, and is likely used as a component in agent systems to... -
4
MiniMax-01
Large-language-model & vision-language-model based on Linear Attention
MiniMax-01 is the official repository for two flagship models: MiniMax-Text-01, a long-context language model, and MiniMax-VL-01, a vision-language model built on top of it. MiniMax-Text-01 uses a hybrid attention architecture that blends Lightning Attention, standard softmax attention, and Mixture-of-Experts (MoE) routing to achieve both high throughput and long-context reasoning. It has 456 billion total parameters with 45.9 billion activated per token and is trained with advanced parallel strategies such as LASP+, varlen ring attention, and Expert Tensor Parallelism, enabling a training context of 1 million tokens and up to 4 million tokens at inference. ... -
Earn up to 16% annual interest with Nexo.Generate interest, borrow against your crypto, and trade a range of cryptocurrencies — all in one platform. Geographic restrictions, eligibility, and terms apply.
-
5
Qwen3-Omni
Qwen3-omni is a natively end-to-end, omni-modal LLM
Qwen3-Omni is a natively end-to-end multilingual omni-modal foundation model that processes text, images, audio, and video and delivers real-time streaming responses in text and natural speech. It uses a Thinker-Talker architecture with a Mixture-of-Experts (MoE) design, early text-first pretraining, and mixed multimodal training to support strong performance across all modalities without sacrificing text or image quality. The model supports 119 text languages, 19 speech input languages, and 10 speech output languages. It achieves state-of-the-art results: across 36 audio and audio-visual benchmarks, it hits open-source SOTA on 32 and overall SOTA on 22, outperforming or matching strong closed-source models such as Gemini-2.5 Pro and GPT-4o. ... -
6
pomegranate
Fast, flexible and easy to use probabilistic modelling in Python
pomegranate is a library for probabilistic modeling defined by its modular implementation and treatment of all models as the probability distributions they are. The modular implementation allows one to easily drop normal distributions into a mixture model to create a Gaussian mixture model just as easily as dropping a gamma and a Poisson distribution into a mixture model to create a heterogeneous mixture. But that's not all! Because each model is treated as a probability distribution, Bayesian networks can be dropped into a mixture just as easily as a normal distribution, and hidden Markov models can be dropped into Bayes classifiers to make a classifier over sequences. ... -
7
Grok-1
Open-source, high-performance Mixture-of-Experts large language model
Grok-1 is a 314-billion-parameter Mixture-of-Experts (MoE) large language model developed by xAI. Designed to optimize computational efficiency, it activates only 25% of its weights for each input token. In March 2024, xAI released Grok-1's model weights and architecture under the Apache 2.0 license, making them openly accessible to developers. The accompanying GitHub repository provides JAX example code for loading and running the model. -
8
L5RCM
GM tool for the RPG game "Legend of the Five Rings" 4th edition
...It can be used to create PC and NPC and to manage characters during playtime. The Legend of the Five Rings RPG is a role playing game that takes place in Rokugan, a mixture of Asian medieval culture. You can find more on L5R RPG on its official site. -
9
Mixtral offloading
Run Mixtral-8x7B models in Colab or consumer desktops
Mixtral-Offloading is an open-source project designed to enable efficient inference of large Mixture-of-Experts language models such as Mixtral-8x7B on hardware with limited GPU memory. 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. ... -
Custom VMs From 1 to 96 vCPUs With 99.95% UptimeLive migration and automatic failover keep workloads online through maintenance. One free e2-micro VM every month.
-
10
DeepSeek MoE
Towards Ultimate Expert Specialization in Mixture-of-Experts Language
DeepSeek-MoE (“DeepSeek MoE”) is the DeepSeek open implementation of a Mixture-of-Experts (MoE) model architecture meant to increase parameter efficiency by activating only a subset of “expert” submodules per input. The repository introduces fine-grained expert segmentation and shared expert isolation to improve specialization while controlling compute cost. For example, their MoE variant with 16.4B parameters claims comparable or better performance to standard dense models like DeepSeek 7B or LLaMA2 7B using about 40% of the total compute. ... -
11
LLaMA-MoE
Building Mixture-of-Experts from LLaMA with Continual Pre-training
LLaMA-MoE is an open-source project that builds mixture-of-experts language models from LLaMA through expert partitioning and 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. ... -
12
QuantResearch
Quantitative analysis, strategies and backtests
...These include implementations of factor models, statistical arbitrage strategies, portfolio optimization methods, and reinforcement learning approaches to trading. The repository also explores financial modeling topics such as vector autoregression, Gaussian mixture models, and option pricing techniques. Many notebooks demonstrate backtesting pipelines that allow users to evaluate trading strategies using historical market data. The project integrates machine learning methods with traditional quantitative finance models, illustrating how statistical techniques can be applied to asset management and trading. -
13
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
-
14
Twinify
Privacy-preserving generation of a synthetic twin to a data set
twinify is a software package for the privacy-preserving generation of a synthetic twin to a given sensitive tabular data set. On a high level, twinify follows the differentially private data-sharing process introduced by Jälkö et al.. Depending on the nature of your data, twinify implements either the NAPSU-MQ approach described by Räisä et al. or finds an approximate parameter posterior for any probabilistic model you formulated using differentially private variational inference (DPVI).... -
15
Scripting Language Bindings
A port of WFOPT to the several scripting languages
This project contains bindings for various scripting languages to the Wheefun Options Parsing Library. It is meant to provide parity with the C implementation so .NET languages can take advantage of WFOPT. For more information, please see the main page. -
16
SVoice (Speech Voice Separation)
We provide a PyTorch implementation of the paper Voice Separation
...The model employs gated neural networks with recurrent processing blocks that disentangle voices over multiple computational steps, while maintaining speaker consistency across output channels. Separate models are trained for different speaker counts, and the largest-capacity model dynamically determines the actual number of speakers in a mixture. The repository includes all necessary scripts for training, dataset preparation, distributed training, evaluation, and audio separation. -
17
Polluted gases Estimator
Create, estimate and analyses gas pollutants and the possible costs
The SPS machines are not designed to filter the particulates to be of acceptable size, level or concentration, unlike Wet Scrubbers, Electrostatic Precipitators, Diesel Particulate Filter DPF etc. The reality is that finest and ultra-finest Particulates tends to behave like gas rather than a gas-solids mixture or simply can be transported by vapour. Is the reason why many filtering technologies are ineffective. However, SPS technology is post-combustion system or last stage of production, which results in packaged CO2 or any other gases, without worrying about gas dilution and intrusion processes which more complicated and costly also are not well investigated, that is a setback for many CO2 technology. ... -
18
Mixture100
Simply percent calculator
For making various mixtures, like homemade alcohol, brine or e-liquids. Add a components, set a percentage in total, set how much I want a given mixture and save, share or print recipe. -
19
fitGCP
Fitting genome coverage distributions with mixture models
...fitGCP is a framework for fitting mixtures of probability distributions to genome coverage profiles. Besides commonly used distributions, fitGCP uses distributions tailored to account for common artifacts. The mixture models are iteratively fitted based on the Expectation-Maximization algorithm. Please find the accompanying paper here: http://dx.doi.org/10.1093/bioinformatics/btt147 -
20
de:Code
Top-down dungeon crawler
de:Code is a 3rd-person top-down dungeon crawler. The world is procedurally generated using the file structure of the users hard drive. The game will use a mixture of different genres including steampunk, fantasy, mid-evil, and modern. The user will have to travel down 4 main paths each progressively harder than the last and each will have more than one genre conflicting inside. Each main path (connected by a central hub) will get harder as the user progresses down, finally reaching a unique boss at the end. -
21
GalGen Trader
A 2D Elite-style game written in Python 3.2.
...This functionality is supported by the game being written in Python 3, with its powerful support for including external code at run-time. The graphics will be a simple mixture of 2D raster and vector imagery with text-based menus, and the music will be highly reminiscent of the 80's 'chiptunes' that made up the soundtracks for games on the Amiga and Commodore 64. -
22
This project hosts tools used for analysis of Gaussian Mixture Distributions (GMDs) which are used for statistical signal processing. The tools are libraries for implementing GMD operations and programs used to analyze properties of GMDs.
-
23
kiki the nanobot is a 3-D puzzle game. It is basically a mixture of the games Sokoban and Kula-World.
-
24
mod_npy is alternative way of running Python scripts under Apache - alternative to having mod_python installed, or running as CGI. It allows mixing Python code with HTML much like PHP. It's a mixture of Python power and PHP style page writing.
-
25
This project aims at writing a threaded, object-oriented and scriptable game engine targetting the .NET platform, using a mixture of c# or managed c++ ports of famous c/c++ components such as OGRE, CEGUI, Newton, Lua, Python, Audiere, OpenAL, etc...
