Open-source alternatives to Gemma.cpp

llama.cpp is a high-performance C++ inference engine and runtime for executing large language models locally across various hardware architectures. It provides the core components for local model execution, including a dedicated model quantizer for compressing weights into the GGUF format and a system for generating text embeddings for semantic search. The project distinguishes itself through specialized memory and execution optimizations, such as block-wise weight quantization to reduce memory footprints and memory-mapped model loading. It supports structured text generation by using formal

MNN is a high-performance inference engine and framework designed for on-device machine learning. It provides a comprehensive environment for executing, optimizing, and deploying neural network models directly on mobile and resource-constrained edge devices. The framework distinguishes itself through a robust model optimization toolkit that supports quantization, compression, and structural graph manipulation to minimize memory footprint and maximize execution speed. It features a modular architecture that abstracts hardware-specific backends, allowing models to run efficiently across diverse

MiniCPM-V is a multimodal large language model and vision-language system designed for complex visual and linguistic understanding. It functions as an on-device AI model, providing the capacity to process text, images, and video as a compact neural network. The project is specifically developed as an edge AI framework, utilizing quantization and weight sharding to run on memory-constrained mobile chipsets. This allows for the deployment of multimodal intelligence directly on mobile operating systems for local inference. Its capabilities cover multimodal content analysis of high-resolution im

KoboldCPP is a local large language model inference engine and GGUF model runner designed to execute quantized models on personal hardware. It functions as a multimodal AI server and API gateway, providing OpenAI-compatible endpoints that allow third-party clients to interact with locally hosted models. The project distinguishes itself as an AI storytelling backend, featuring dedicated tools for long-form narrative management through persistent memory, world lore tracking, and character state management. It further extends its capabilities as a multimodal server capable of processing text, im

LiteRT-LM is a high-performance inference framework designed to execute large language models locally on mobile, desktop, and IoT hardware. It serves as an on-device model runtime that utilizes CPU, GPU, and NPU acceleration to provide low-latency processing. The framework is distinguished by its ability to process text, vision, and audio inputs through a single multi-modal inference engine. It features a local HTTP server that emulates OpenAI-compatible API endpoints and a WebGPU-based runtime for executing models directly within a web browser. To ensure output reliability, it includes a con

The nexa-sdk is an on-device AI SDK and multimodal inference engine designed to run large language, vision, and audio models locally on mobile and desktop hardware. It functions as a local LLM runtime and NPU acceleration framework, enabling the execution of generative and discriminative models without reliance on cloud services. The project distinguishes itself through a dedicated NPU acceleration framework that optimizes model execution on Neural Processing Units to reduce latency and power consumption. It employs hardware-agnostic backend routing to dynamically distribute computations acro

This project is a library and command-line interface for local large language model inference. It enables the generation of text completions and chat responses from various model architectures. The project provides tools for weight quantization to reduce memory footprints and incorporates hardware acceleration through GPU offloading to increase computation speed. It also includes utilities for model evaluation by measuring perplexity on specific datasets. Capabilities cover the full inference lifecycle, including binary model loading, template-based prompt structuring, and session persistenc

This project is a vision language model framework and vision-to-text pipeline designed for deploying and optimizing models that process both images and text. It provides an on-device inference engine and a vision language model framework to run quantized models locally on mobile and desktop hardware accelerators. The framework features a model quantization toolkit to reduce weight precision for lower memory footprints and increased execution speed on specialized silicon. It also includes an efficient vision encoder utilizing a hybrid encoding system to compress image tokens, which reduces pro

SentencePiece is a text segmentation engine and tokenization library designed for machine learning workflows. It provides a comprehensive toolkit for transforming raw text into subword units or numerical identifiers, enabling consistent data representation for neural network training and inference. The library supports the training of segmentation models from raw text, allowing for the creation of custom vocabularies tailored to specific domain requirements. The project distinguishes itself through its byte-level encoding and fallback mechanisms, which ensure that every input can be represent

This is a webcam-based client for a local llama.cpp server that enables real-time object detection and vision-language model inference directly from a browser. It captures frames from the user's webcam at configurable intervals and sends them to a locally running inference server for analysis, displaying both detection results and textual scene descriptions as they are produced. The application distinguishes itself by combining object detection with vision-language scene description in a single real-time interface, all processed through a local llama.cpp server for private, offline operation.

This project is a comprehensive collection of educational examples and reference implementations for building vision and language models using PyTorch. It serves as a deep learning tutorial covering the end-to-end process of developing neural networks, from initial architecture definition to final production deployment. The repository provides detailed guides on implementing a wide range of domain-specific models, including convolutional neural networks for object detection and segmentation, as well as transformer and recurrent architectures for natural language processing. It emphasizes gene

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

This project is an on-device AI SDK providing a framework for running large language models, vision models, and speech models locally. It serves as an orchestration layer for local LLM execution, ensuring data privacy and offline availability by utilizing hardware acceleration on the device. The SDK is distinguished by its comprehensive voice and multimodal capabilities, including a coordinated voice pipeline for activity detection, speech-to-text, and text-to-speech synthesis. It also provides a dedicated implementation kit for local retrieval-augmented generation and tools for processing co

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

This library is a web-native engine designed to execute pretrained machine learning models directly within the browser. It functions as a client-side inference framework, enabling developers to run complex neural networks for natural language processing, computer vision, and audio tasks without requiring a backend server or external API calls. The framework distinguishes itself by providing a unified pipeline-based abstraction that handles the entire lifecycle of model execution. It manages the dynamic retrieval of model weights and configurations from remote registries, while simultaneously

whisper.cpp is a C++ implementation of the Whisper speech-to-text model, serving as a lightweight machine learning inference engine and quantized runtime. It provides high-performance automatic speech recognition and real-time audio transcription without requiring a Python environment. The project utilizes model quantization to reduce memory usage and increase inference speed on local hardware. It incorporates hardware acceleration to optimize processing speed across different processors. The system covers audio processing capabilities including voice activity detection, speaker diarization,

OpenVINO is an AI inference engine and model serving platform designed to execute optimized deep learning models across CPUs, GPUs, and NPUs through a unified API. It includes a model optimization toolkit for converting, quantizing, and compressing models from various frameworks, alongside a specialized generative AI runtime for large language models. The project distinguishes itself through a plugin-based hardware acceleration layer that maps neural network operations to vendor-specific drivers. It features advanced execution mechanisms such as continuous batching, speculative decoding, and

Moonshine is a complete on-device voice interface toolkit that provides speech recognition, text-to-speech synthesis, phonetic processing, speaker diarization, and intent recognition, all running locally on edge hardware without any cloud dependency. It executes quantized neural networks for speech and language tasks directly on the device, enabling fully offline conversational AI capabilities. The toolkit distinguishes itself by orchestrating multi-turn spoken exchanges through a conversational flow manager that maintains context across interactions and manages branching dialog flows. It inc

tiny-dnn is a header-only C++14 deep learning framework for building, training, and running inference on neural networks. It constructs static computational graphs at compile time using template-based layer composition, with a gradient-based backpropagation engine and minibatch stochastic gradient descent for training, all without external dependencies beyond the C++14 standard library. The framework supports importing pre-trained models from the Caffe framework directly, parsing its binary serialization format without requiring external protocol buffer libraries. It provides CPU-optimized te

Mamba is a deep learning framework designed for building and training sequence models that process long-range data dependencies with linear-time computational efficiency. By utilizing selective state space modeling, the library enables the construction of neural network architectures that replace traditional attention mechanisms with high-performance state space operations. The framework distinguishes itself through the use of data-dependent state gating, which allows the model to dynamically filter information flow based on the input sequence. To ensure high throughput, it incorporates hardw

This project is a self-hosted meeting transcription and summarization tool that converts audio recordings into text transcripts and structured notes using large language models. It functions as an enterprise meeting documentation manager, allowing for the organization and editing of timestamped records. The system prioritizes data privacy through local-first processing and the ability to deploy on private infrastructure. It supports a provider-agnostic architecture, enabling users to connect to local AI engines, self-hosted servers, or cloud-based API endpoints for both transcription and summ

Caffe is a high-performance deep learning framework designed for training and deploying deep neural networks. It functions as a machine learning engine and a convolutional neural network library, providing a C++ backend to accelerate computations on both GPUs and CPUs. The system includes a specialized toolset for computer vision, enabling tasks such as object detection, semantic segmentation, and large-scale image retrieval. It supports the deployment of pre-trained models for image and scene recognition, as well as the ability to fine-tune neural network weights for specialized tasks. The

This project is a training pipeline and framework for developing Chinese language models based on the Llama 2 architecture. It functions as a distributed GPU trainer and dataset preprocessing toolkit designed for both the initial pre-training of baseline models and subsequent supervised fine-tuning. The system distinguishes itself through a specialized workflow for Chinese text, incorporating a data curation pipeline that uses similarity hashing for deduplication and a tokenization process that converts raw text into memory-mapped binary files for efficient disk access. It implements a superv

This project is an educational platform and research toolkit designed to teach deep learning through a combination of mathematical theory, visual diagrams, and executable code. It provides a comprehensive environment for building, training, and evaluating neural networks, grounding complex concepts in interactive computational notebooks that allow for hands-on experimentation. The framework distinguishes itself by interleaving theoretical foundations—including linear algebra, calculus, and probability—with practical implementations across multiple industry-standard libraries. It supports flex

This project is a comprehensive, community-driven repository that serves as a centralized catalog for computer vision research and development. It functions as a structured index of academic papers, open-source software libraries, public datasets, and educational tutorials, providing a navigation point for the complex landscape of modern vision technology. The repository distinguishes itself through a taxonomy-based indexing system that maps the relationships between foundational research, influential academic figures, and their corresponding software implementations. By utilizing a lightweig

gpt-neox is a distributed training system and framework for building large-scale autoregressive language models. It implements the transformer architecture and provides a toolkit for training models with billions of parameters by distributing weights across compute clusters. The framework distinguishes itself through extensive support for distributed model parallelism, including pipeline and sequence parallelism, to overcome single-device memory limits. It further supports sparse model architectures using a mixture of experts system with Sinkhorn-based routing. The project covers a broad ran