Search Results for "compact" - Page 2
Sort By:
Showing 50 open source projects for "compact"
View related business solutions-
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
Atera - an All-in-one platform for IT managementYour IT essentials, integrated & elevated. Take your IT management from automated to autonomous, download Atera's agent to start your free trial!
-
1
HY-MT
Hunyuan Translation Model Version 1.5
HY-MT (Hunyuan Translation) is a high-quality multilingual machine translation model suite developed to support mutual translation across dozens of languages with strong performance even at smaller model scales. It ships with both an 1.8 B parameter model and a larger 7 B model, the latter optimized not only for direct translation but also for formatted and contextualized output, allowing better handling of terminology and mixed-language content. The project emphasizes both speed and... -
2
nanocode
Minimal Claude Code alternative. Single Python file, zero dependencies
nanocode is a minimalist coding agent implementation designed as a compact alternative to Claude Code, packaged in a single Python file with no external dependencies and totaling around 250 lines of code. It implements a full agentic loop where the model can reason, decide when to use tools, execute those tools, and iterate until producing a final answer, making it useful for simple AI-assisted coding workflows. -
3
Magika
Fast and accurate AI powered file content types detection
Magika is an AI-powered file-type detector that uses a compact deep-learning model to classify binary and textual files with high accuracy and very low latency. The model is engineered to be only a few megabytes and to run quickly even on CPU-only systems, making it practical for desktop apps, servers, and security pipelines. Magika ships as a command-line tool and a library, providing drop-in detection that improves on traditional “magic number” and heuristic approaches, especially for ambiguous or short files. ... -
4
LEANN
Local RAG engine for private multimodal knowledge search on devices
...LEANN introduces a storage-efficient approximate nearest neighbor index combined with on-the-fly embedding recomputation to avoid storing large embedding vectors. By recomputing embeddings during queries and using compact graph-based indexing structures, LEANN can maintain high search accuracy while minimizing disk usage. It aims to act as a unified personal knowledge layer that connects different types of data such as documents, code, images, and other local files into a searchable context for language models. -
Auth0 B2B Essentials: SSO, MFA, and RBAC Built InAuth0's B2B Essentials plan gives you everything you need to ship secure multi-tenant apps. Unlimited orgs, enterprise SSO, RBAC, audit log streaming, and higher auth and API limits included. Add on M2M tokens, enterprise MFA, or additional SSO connections as you scale.
-
5
Gitingest
Create prompt-friendly codebase digests from any Git repository URL
...It analyzes a repository and produces a consolidated textual representation that includes the file structure and code content in an organized format. This makes it easier to provide meaningful code context when working with AI systems that require compact, readable inputs. Developers can generate these digests from either a local directory or a remote repository by supplying a repository path or URL. The generated output is optimized for prompt usage, helping AI models understand codebases more effectively without requiring manual file aggregation. In addition to producing the code digest, Gitingest also calculates statistics about the extracted content such as repository structure, total size of the extract, and token count. ... -
6
Machine learning algorithms
Minimal and clean examples of machine learning algorithms
...The repository includes implementations of both supervised and unsupervised learning techniques, along with dimensionality reduction and clustering methods. Many of the algorithms are written in a simplified style that prioritizes clarity and educational value over production-level optimization. Because the code is compact and easy to follow, it is often used as a learning resource by developers who want to understand how machine learning algorithms are constructed. -
7
Chat with LLMs Everywhere
Run PyTorch LLMs locally on servers, desktop and mobile
TorchChat is an open-source project from the PyTorch ecosystem designed to demonstrate how large language models can be executed efficiently across different computing environments. The project provides a compact codebase that illustrates how to run conversational AI systems using PyTorch models on laptops, servers, and mobile devices. It is intended primarily as a reference implementation that shows developers how to integrate large language models into applications without requiring a large or complex infrastructure stack. TorchChat supports running models through Python interfaces as well as integrating them directly into native applications written in languages such as C or C++. ... -
8
Vision Transformer Pytorch
Implementation of Vision Transformer, a simple way to achieve SOTA
...It breaks down the model into patch embedding, positional encoding, multi-head self-attention, feed-forward blocks, and a classification head so you can understand each component in isolation. The code is intentionally compact and modular, which makes it easy to tinker with hyperparameters, depth, width, and attention dimensions. Because it stays close to vanilla PyTorch, you can integrate custom datasets and training loops without framework lock-in. It’s widely used as an educational reference for people learning transformers in vision and as a lightweight baseline for research prototypes. ... -
9
llama2.c
Inference Llama 2 in one file of pure C
...While it can technically load Meta’s official Llama 2 models, current support is limited to fp32 precision, meaning practical use is capped at models up to around 7B parameters. The goal of llama2.c is to demonstrate how a compact and transparent implementation can perform meaningful inference even with small models, emphasizing simplicity, clarity, and accessibility. The project builds upon lessons from nanoGPT and takes inspiration from llama.cpp, focusing instead on minimalism and educational value over large-scale performance. -
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.
-
10
Rhino
On-device Speech-to-Intent engine powered by deep learning
Rhino is Picovoice's Speech-to-Intent engine. It directly infers intent from spoken commands within a given context of interest, in real-time. The end-to-end platform for embedding private voice AI into any software in a few lines of code. Design with no limits on top of a modular platform. Create use-case-specific voice AI models in seconds. Develop voice features with a few lines of code using intuitive and cross-platform SDKs. Deliver voice AI everywhere: on-device, mobile, web browsers,... -
11
DeiT (Data-efficient Image Transformers)
Official DeiT repository
...Its key idea is a specialized distillation strategy—including a learnable “distillation token”—that lets a transformer learn effectively from a CNN or transformer teacher on modest-scale datasets. The project provides compact ViT variants (Tiny/Small/Base) that achieve excellent accuracy–throughput trade-offs, making transformers practical beyond massive pretraining regimes. Training involves carefully tuned augmentations, regularization, and optimization schedules to stabilize learning and improve sample efficiency. The repo offers pretrained checkpoints, reference scripts, and ablation studies that clarify which ingredients matter most for data-efficient ViT training. -
12
AudioBC
Offline desktop app to convert EPUB to MP3 using Kokoro-82M neural TTS
...It is built with a focus on privacy and simplicity, offering a streamlined three-step workflow: Extract, Edit, and Convert. Key Features: Neural Quality TTS: Uses the compact yet powerful Kokoro-82M model for high-fidelity, expressive voice synthesis. Privacy-First & Offline: After a one-time initial model download, all processing happens on your CPU. Your books never leave your computer. Multi-Language Support: Curated voices for English (US & UK), Italian, French, Spanish, and Portuguese (BR). ... -
13
ConsistencyDecoder
Consistency Distilled Diff VAE
...It demonstrates how consistency models can enhance visual fidelity while maintaining efficiency, reducing artifacts common in GAN-decoded outputs. The repository includes installation instructions, usage examples, and visual comparisons to highlight improvements. Though compact, it provides an accessible entry point for experimenting with advanced decoding strategies in diffusion-based generative models. -
14
picoGPT
An unnecessarily tiny implementation of GPT-2 in NumPy
picoGPT is a minimal implementation of the GPT-2 language model designed to demonstrate how transformer-based language models work at a conceptual level. The repository focuses on educational clarity rather than production performance, implementing the core components of the GPT architecture in a concise and readable way. It allows users to understand how tokenization, transformer layers, attention mechanisms, and autoregressive text generation operate in modern large language models. The... -
15
minGPT
A minimal PyTorch re-implementation of the OpenAI GPT
minGPT is a minimalist, educational re-implementation of the GPT (Generative Pretrained Transformer) architecture built in PyTorch, designed by Andrej Karpathy to expose the core structure of a transformer-based language model in as few lines of code as possible. It strips away extraneous bells and whistles, aiming to show how a sequence of token indices is fed into a stack of transformer blocks and then decoded into the next token probabilities, with both training and inference supported.... -
16
WaveRNN
WaveRNN Vocoder + TTS
WaveRNN is a PyTorch implementation of DeepMind’s WaveRNN vocoder, bundled with a Tacotron-style TTS front end to form a complete text-to-speech stack. As a vocoder, WaveRNN models raw audio with a compact recurrent neural network that can generate high-quality waveforms more efficiently than many traditional autoregressive models. The repository includes scripts and code for preprocessing datasets such as LJSpeech, training Tacotron to produce mel spectrograms, training WaveRNN on those spectrograms (with optional GTA data), and finally generating audio. ... -
17
Catalyst
Accelerated deep learning R&D
Catalyst is a PyTorch framework for accelerated Deep Learning research and development. It allows you to write compact but full-featured Deep Learning pipelines with just a few lines of code. With Catalyst you get a full set of features including a training loop with metrics, model checkpointing and more, all without the boilerplate. Catalyst is focused on reproducibility, rapid experimentation, and codebase reuse so you can break the cycle of writing another regular train loop and make something totally new. ... -
18
Old Photo Restoration
Bringing Old Photo Back to Life (CVPR 2020 oral)
...And the translation between these two latent spaces is learned with synthetic paired data. This translation generalizes well to real photos because the domain gap is closed in the compact latent space. Besides, to address multiple degradations mixed in one old photo, we design a global branch with a partial nonlocal block targeting to the structured defects, such as scratches and dust spots. -
19
micrograd
A tiny scalar-valued autograd engine and a neural net library
...On top of the core autograd “Value” concept, the project includes a small neural network library that lets you define and train simple models with a PyTorch-like feel, including multilayer perceptrons. The repository is intentionally compact and readable, prioritizing clarity over performance so learners can follow every step of gradient flow and parameter updates. It is commonly used as a learning bridge between basic calculus intuition and full-scale deep learning frameworks, helping developers understand why autodiff libraries behave the way they do. -
20
VGGFace2
VGGFace2 Dataset for Face Recognition
...Alongside the dataset, the repository provides pre-trained models based on ResNet-50 and SE-ResNet-50 architectures, trained with both MS-Celeb-1M pretraining and fine-tuning on VGGFace2. These models achieve strong verification performance on benchmarks such as IJB-B and include variants with lower-dimensional embeddings for compact feature representation. The project also includes preprocessing tools, face detection scripts, and etc. -
21
DeepSDF
Learning Continuous Signed Distance Functions for Shape Representation
DeepSDF is a deep learning framework for continuous 3D shape representation using Signed Distance Functions (SDFs), as presented in the CVPR 2019 paper DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation by Park et al. The framework learns a continuous implicit function that maps 3D coordinates to their corresponding signed distances from object surfaces, allowing compact, high-fidelity shape modeling. Unlike traditional discrete voxel grids or meshes, DeepSDF encodes shapes as continuous neural representations that can be smoothly interpolated and used for reconstruction, generation, and analysis. The repository provides complete tooling for preprocessing mesh datasets (e.g., ShapeNet), training DeepSDF models, reconstructing meshes from learned latent codes, and quantitatively evaluating results with metrics such as Chamfer Distance and Earth Mover’s Distance. -
22
Resemblyzer
A python package to analyze and compare voices with deep learning
Resemblyzer is a Python package for analyzing and comparing voices with deep learning. It works by turning speech audio into a compact voice embedding that represents the speaker’s vocal characteristics. These embeddings can then be used for speaker similarity, clustering, diarization experiments, voice comparison, and audio dataset exploration. The project is useful for researchers and developers who need a practical way to reason about speaker identity without building a voice encoder from scratch. ... -
23
InferSent
InferSent sentence embeddings
...Trained on large NLI datasets, the embeddings generalize across tasks like sentiment analysis, entailment, paraphrase detection, and semantic similarity with simple linear classifiers. The repository provides pretrained vectors, training scripts, and clear examples for evaluating transfer on a wide suite of benchmarks. Because the encoder is compact and language-agnostic at the interface level, it’s easy to drop into production pipelines that need robust semantic features. InferSent helped popularize the idea that supervised objectives (like NLI) can yield strong general-purpose sentence encoders, and it remains a reliable baseline against which to compare newer models. -
24
Siamese and triplet learning
Siamese and triplet networks with online triplet mining in PyTorch
Siamese and triplet learning is a PyTorch implementation of Siamese and triplet neural network architectures designed for learning embedding representations in machine learning tasks. These types of networks learn to map images into a compact feature space where the distance between vectors reflects the similarity between inputs. Such embeddings are commonly used in applications like face recognition, image similarity search, and few-shot learning. The repository demonstrates how to train these models using contrastive loss and triplet loss functions, which encourage embeddings of similar samples to be close while pushing dissimilar samples farther apart. ... -
25
gpt-oss-20b
OpenAI’s compact 20B open model for fast, agentic, and local use
GPT-OSS-20B is OpenAI’s smaller, open-weight language model optimized for low-latency, agentic tasks, and local deployment. With 21B total parameters and 3.6B active parameters (MoE), it fits within 16GB of memory thanks to native MXFP4 quantization. Designed for high-performance reasoning, it supports Harmony response format, function calling, web browsing, and code execution. Like its larger sibling (gpt-oss-120b), it offers adjustable reasoning depth and full chain-of-thought visibility...
