Awesome GitHub RepositoriesQuantized Model Implementations

Grok-1 is an open-weights large language model implementation featuring a sparse mixture-of-experts architecture. It is designed for high-performance text generation and natural language processing by activating only a subset of specialized expert layers per token. The model utilizes 8-bit weight quantization to reduce memory overhead and accelerate loading. To manage its high parameter count, the implementation supports activation sharding, which distributes the memory load across multiple hardware devices during execution. The project covers large-scale model inference, including text comp

ChatGLM-6B is an open-source bilingual large language model designed for natural dialogue and text generation in both English and Chinese. It is structured as a dialogue model capable of tasks such as role-playing and information extraction. The project provides implementations for quantized language models, using low-precision weights to reduce GPU memory requirements for local inference. It also supports parameter-efficient fine-tuning, allowing model behavior to be optimized for specific tasks without requiring full retraining. The model includes capabilities for local execution on GPUs a

fastText is a library and framework for word embedding generation, text vectorization, and supervised text classification. It provides tools to transform raw text into fixed-length vector representations and to train models that assign category labels to sentences or documents. The system utilizes subword-based vectorization and character n-gram embeddings, allowing it to generate meaningful vectors for words that were not present during training. To manage resource usage, it includes a quantized language model implementation that employs product quantization and dimensionality reduction to d

Qwen-7B is a pretrained causal language model designed for natural language generation, text processing, and complex reasoning tasks. It is available as an instruction-tuned model optimized for conversational interactions and a tool-use model capable of executing function calls and interacting with external APIs. The project provides a quantized version of the model to reduce GPU memory usage and supports the development of autonomous agents that can execute code and perform functions to complete complex goals. The system covers a wide range of capabilities including model fine-tuning throug

ChatGLM2-6B is an open-weight large language model designed for natural language conversations and text generation in both English and Chinese. It functions as a bilingual chat model capable of processing and maintaining coherence across text sequences up to 32K tokens. The model is optimized for local deployment through precision quantization, which reduces memory requirements to allow execution on consumer-grade hardware. It supports distributing model weights across multiple graphics cards to handle parameters that exceed the memory of a single device. The project covers capabilities for

ChatGLM2-6B is a bilingual chat large language model designed for natural conversation and text generation in both English and Chinese. It functions as a fine-tunable language model that supports updating weights via specialized scripts to adapt to specific datasets and tasks. The project serves as a quantized inference engine and multi-GPU model orchestrator, enabling the execution of large models on consumer-grade hardware. It is capable of processing long context sequences up to 32K tokens to maintain understanding across extended documents. The system covers capabilities for multilingual

ChatGLM3 is an open-weights large language model designed for bilingual conversational interactions in English and Chinese. It functions as a tool-augmented system capable of calling external functions and executing internal code to resolve complex tasks. The model utilizes four-bit quantization to reduce memory requirements, enabling inference on consumer hardware and diverse processing units including GPUs and CPUs. It features an expanded context window for processing and summarizing long documents and includes a supervised fine-tuning pipeline for adapting the model to specialized domains

MiniCPM is a collection of small language models designed for local, on-device deployment in resource-constrained environments. The project focuses on running dense Transformer models on consumer hardware, including GPUs, CPUs, and Apple Silicon, without requiring custom code forks. The project distinguishes itself through heavy optimization for edge hardware, utilizing quantized weight compression in GGUF and MLX formats to reduce memory overhead. It implements advanced inference techniques such as speculative sampling and radix-tree prefix caching to accelerate generation speed and throughp

Intel XPU LLM Acceleration Library is a toolkit designed to accelerate large language model inference and finetuning on Intel CPUs, GPUs, and NPUs. It provides a distributed inference engine for scaling models across multiple accelerators, a multimodal model runtime for vision and speech tasks, and a low-bit model quantization tool for converting weights into INT4, FP8, and GGUF formats. The project features a parameter-efficient finetuning framework that enables model adaptation using QLoRA and DPO on Intel hardware. It distinguishes itself by providing specialized optimizations for Intel XP

Dieses Projekt bietet eine Suite leichtgewichtiger Gesichtserkennungsmodelle, die für die Inferenz mit hoher Geschwindigkeit auf Edge-Computing-Geräten entwickelt wurden. Es basiert auf einer kompakten neuronalen Netzwerkarchitektur, die die Erkennung menschlicher Gesichter in Umgebungen mit begrenzten Rechenressourcen und Leistungsbeschränkungen ermöglicht. Das System bietet quantisierte Gesichtserkenner in mehreren Formaten, um die Kompatibilität über verschiedene Hardwarearchitekturen hinweg sicherzustellen. Es enthält Dienstprogramme für den Modellexport und die Quantisierung, wodurch trainierte Gewichte in standardisierte Formate für hardwareunabhängige Bereitstellungen konvertiert werden können. Das Projekt deckt Workflows für das Training benutzerdefinierter Modelle und die Verarbeitung von Bilddatensätzen ab, um Präzisions- und Geschwindigkeitsprofile anzupassen. Es unterstützt Echtzeit-Computer-Vision-Aufgaben durch die Verwendung von Integer-only-Arithmetik und einer C-basierten Inferenz-Runtime, um den Overhead auf eingebetteter Hardware zu minimieren.

CogVLM is a multimodal large language model designed for visual reasoning and multi-turn dialogue. It functions as a visual grounding model and a quantized vision model, combining text and image processing to perform complex understanding and maintain context across visual inputs. The project includes capabilities as a GUI automation agent, allowing it to analyze application screenshots, plan operational steps, and return precise screen coordinates for interface interaction. It further supports visual grounding by generating bounding box coordinates to map text descriptions to specific spatia

mistral.rs is an inference engine for large language models that runs locally and exposes models behind OpenAI and Anthropic-compatible APIs. It serves as a multi-model serving platform, capable of loading several models in a single server process with per-request routing and on-demand loading and unloading. The engine supports multimodal inference, processing text alongside images, video, audio, and speech inputs, and includes a quantized model deployment runtime that reduces memory use and speeds up inference on consumer hardware. The project distinguishes itself through an agentic tool exe

YOLO-World is a vision-language framework and open-vocabulary object detection model. It identifies objects in images and video based on free-form text prompts without requiring predefined category labels. The system enables the identification of arbitrary objects by fusing image features with text embeddings. It includes a specialized tool for automated image labeling, which generates bounding box annotations for custom datasets using text-based prompts. The project provides a deployment pipeline for converting models into quantized ONNX and TFLite formats, supporting real-time inference on

Gluon-CV ist eine MXNet-Computer-Vision-Bibliothek, die eine umfassende Sammlung vortrainierter Vision-Architekturen und Trainings-Pipelines bereitstellt. Sie dient als Deep-Learning-Research-Toolkit und Model-Zoo mit State-of-the-Art-Gewichten für die Bild- und Videoanalyse. Das Projekt enthält eine spezialisierte Bibliothek für Human-Pose-Estimation sowie ein Toolkit zur Modellkompression. Diese Tools ermöglichen das Pruning und die Quantisierung von Deep-Learning-Modellen, um die Inferenzgeschwindigkeit zu erhöhen und die Bereitstellung auf ressourcenbeschränkter Edge-Hardware zu erleichtern. Die Bibliothek deckt ein breites Spektrum an Vision-Funktionen ab, darunter Bildklassifizierung, Objekterkennung sowie semantische und Instanz-Segmentierung. Sie bietet zudem Tools für die Videoanalyse, wie Action-Recognition, Objekt-Tracking und monokulare Tiefenschätzung. Das Training wird durch automatisierte Pipelines und verteilte Multi-GPU-Workloads unterstützt, um die Modellkonvergenz zu beschleunigen.

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a config-driven system for instantiating components, orchestrating distributed training, and managing parameter-efficient fine-tuning with quantization support, all through YAML-based configurations and command-line overrides. The library distinguishes itself through its comprehensive post-training workflow orchestration, combining supervised fine-tuning, preference optimization (DPO, PPO, GRPO), knowledge distillation, and quantization-aware training in a single configurable pip

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a configurable training pipeline orchestrated through YAML recipes, with CLI overrides and component swapping, distributed training via FSDP2, memory optimizations, and parameter-efficient fine-tuning methods like LoRA, DoRA, and QLoRA. The library distinguishes itself through its YAML-driven configuration system that defines all training parameters and instantiates components from config files, with full CLI override capability for any field or component at launch time. It suppo

AutoGPTQ ist ein Framework zur Modellkomprimierung, das entwickelt wurde, um den Speicherbedarf zu reduzieren und die Inferenzgeschwindigkeit großer Sprachmodelle zu erhöhen. Es nutzt den GPTQ-Algorithmus zur Komprimierung von Modellgewichten, wodurch diese Modelle auf Hardware mit begrenztem VRAM ausgeführt werden können. Das Toolkit bietet eine Architektur-Quantisierungspipeline, die die Integration benutzerdefinierter Modellklassen für verschiedene neuronale Netzwerkarchitekturen unterstützt. Es enthält eine Mixed-Precision-Inferenz-Engine mit optimierten Kernels, um die Matrixmultiplikation während des Deployments zu beschleunigen. Das Framework deckt den gesamten Workflow der Gewichtskomprimierung ab, von der Kalibrierung und Quantisierung bis hin zur Genauigkeitsbewertung nachgelagerter Aufgaben. Diese Tools messen den Performance-Verlust durch den Vergleich der Ausgaben quantisierter Modelle mit den Originalgewichten bei Benchmark-Aufgaben.

CTranslate2 is a C++ inference engine and runtime for Transformer models, designed to execute models on both CPU and GPU with optimizations for speed and memory efficiency. It functions as a model format converter, quantization tool, and REST API server, enabling deployment of neural machine translation, automatic speech recognition, and text generation models. The engine distinguishes itself through a suite of runtime optimizations including layer fusion, weight-matrix quantization, batch-by-length grouping, and a caching allocator that reuses GPU memory. It supports tensor-parallel model di

FastVideo is a comprehensive system for accelerated video generation, serving as a video generation inference engine, a video diffusion training framework, and a modular pipeline orchestrator. It provides a distributed transformer optimizer and a distillation toolkit designed to reduce denoising steps and model complexity to increase frame rates. The project distinguishes itself through specialized acceleration techniques, including joint distillation and sparse attention training. It implements low-step video generation and weight quantization to FP8 or FP4 precision to increase throughput a