28 repositorios
Systems that convert diverse data types into unified numerical sequences for latent space processing.
Distinguishing note: Focuses on the alignment of text and visual tokens into a shared latent space, distinct from standard NLP tokenization.
Explore 28 awesome GitHub repositories matching artificial intelligence & ml · Multi-Modal Tokenizers. Refine with filters or upvote what's useful.
CLIP is a neural network architecture designed to map visual and textual data into a shared latent vector space. By utilizing transformer-based feature extraction and multi-modal tokenization, the system aligns images and natural language strings, enabling cross-modal similarity analysis and semantic classification. The project functions as a zero-shot classification engine, identifying image content by calculating the cosine similarity between visual features and arbitrary text labels without requiring task-specific retraining. Beyond inference, it serves as a research toolkit for evaluating
Converts natural language strings into numerical sequences that align with visual features within a unified latent representation space.
This project is a comprehensive framework for building AI-powered applications, providing a unified toolkit for orchestrating language models, autonomous agents, and interactive user interfaces. It serves as a central library for managing the entire lifecycle of AI interactions, from initial prompt generation and model provider abstraction to complex, multi-step reasoning and tool execution. The framework distinguishes itself through its deep integration with frontend development, specifically by enabling generative user interfaces that render dynamic components directly from model outputs. I
Converts complex tool outputs like images or files into standardized formats that models can interpret and incorporate into their responses.
Janus is a multimodal large language model and unified framework that integrates visual understanding and image generation within a single neural network. It functions as both a visual understanding model for analyzing images and a text-to-image generator. The system uses a unified transformer backbone and a multimodal latent space to bridge the gap between text and visual data. This architecture employs decoupled visual encoding and cross-modal tokenization to separate the paths for discriminative understanding and generative tasks, representing images as grids of discrete codes. The projec
Employs multi-modal tokenizers to convert images into a discrete sequence of tokens shared with text.
Clip-as-service is a deployable framework for generating multi-modal embeddings and executing neural searches. It provides a vector embedding server and a CLIP embedding API to convert images and text into shared vector representations via network interfaces. The system functions as a multi-modal ranking system and neural search engine, enabling the retrieval of images through text queries or the identification of matching text descriptions for images. It also includes a visual reasoning service used to analyze images and verify object presence, counts, and colors by comparing visual data aga
Converts images and text into vector representations via network requests for neural search and similarity tasks.
Este proyecto es un servicio de incrustación BERT de alto rendimiento y servidor de inferencia diseñado para mapear secuencias de texto en vectores numéricos de longitud fija. Funciona como un microservicio de aprendizaje automático y servidor de modelos distribuido que desacopla el manejo de solicitudes de la computación pesada. El sistema utiliza una infraestructura de mensajería ZeroMQ para proporcionar comunicación de baja latencia entre clientes distribuidos y el servidor de inferencia. Incorpora procesamiento por lotes del lado del servidor y escalado de carga de trabajo de GPU para maximizar la utilización del hardware y gestionar grandes volúmenes de solicitudes. La plataforma admite infraestructura de búsqueda semántica generando incrustaciones intermodales tanto para texto como para imágenes dentro de un espacio vectorial compartido. Esto permite la búsqueda intermodal, la clasificación de relevancia de contenido y la re-clasificación de resultados basada en la alineación semántica entre el contenido visual y las descripciones de texto. El servicio se puede implementar como un microservicio elástico accesible a través de protocolos gRPC, HTTP o WebSocket, con streaming dúplex sin bloqueo para manejar grandes conjuntos de datos.
Converts text and images into a unified high-dimensional vector space for cross-modal retrieval.
InstantID is a diffusion-based identity preservation framework designed for zero-shot image generation. It allows for the synthesis of images featuring a specific person's facial identity using a single reference photo without requiring additional model training or fine-tuning. The project distinguishes itself through the use of consistency model distillation to accelerate inference, reducing the number of steps needed to produce high-quality results. It combines identity-preserving feature extraction with multi-modal prompt integration to merge visual embeddings from a reference image with t
Combines textual descriptions with visual identity embeddings to control the final output of the diffusion process.
PyOD is a Python anomaly detection library used to identify outliers in tabular, time series, graph, text, and image data. It provides a collection of algorithms for detecting anomalous data points and includes a unified detector interface that standardizes input and output signatures across its available detection algorithms. The project features a multi-modal outlier detector for identifying anomalies across diverse formats including unstructured text and images, as well as a specialized toolkit for graph-based and time-series anomaly detection. It includes an ensemble framework for combini
Transforms diverse data formats into a common representation for unified outlier scoring.
Moshi is a real-time voice foundation model and speech-to-speech framework designed for bidirectional, low-latency conversations. It functions as a full-duplex voice interface that processes audio and text concurrently in a single stream, enabling natural human-machine dialogue without sequential processing delays. The system utilizes a neural audio codec to compress high-fidelity audio into low-bitrate tokens for efficient transmission. To manage complex responses and reasoning, it employs internal monologue modeling, which generates a hidden stream of thought tokens alongside audible speech
Converts audio and text into a unified token stream for single-sequence processing by the model.
ImageBind is a multi-modal embedding model and joint representation learner that maps images, text, audio, and other modalities into a single shared vector space. It functions as a cross-modal retrieval framework designed to bind multiple sensory inputs into one cohesive mathematical embedding. The system uses a contrastive learning architecture to align disparate data types by maximizing the similarity between related samples. This allows the model to perform zero-shot multimodal classification and execute cross-modal data retrieval, such as locating visual content via natural language descr
Maps images, text, audio, and other modalities into a single shared vector space using a neural network.
This repository provides a collection of reference implementations and code examples for training and deploying machine learning models using the MLX framework. It serves as a practical guide for executing distributed training, fine-tuning large language models, converting model weights, and implementing multimodal generative workflows. The project distinguishes itself through specialized examples for local hardware execution, featuring weight quantization to reduce memory usage and low-rank adaptation for parameter-efficient fine-tuning. It also includes scripts for transforming external mod
Implements neural networks that map images and text into a shared vector space for joint retrieval.
lmdeploy is a high-performance inference engine and deployment framework for large language models and vision models. It functions as a multi-modal model server and compression toolkit designed to serve models with high throughput and low latency. The system enables the distribution of model services across multiple machines using request-based load balancing and tensor parallelism. It includes specialized tools for model quantization and compression to reduce the memory footprint of weights and caches. The framework covers broad capability areas including production deployment, distributed
Coordinates the flow of image and text data through distinct encoders before processing them in a unified transformer.
GLM-4 is a large language model and fine-tuning framework designed for human-like text production, complex reasoning, and multilingual conversation. It functions as a multimodal system capable of processing high-resolution visual content and as a long-context model designed to analyze documents with a context window of up to one million tokens. The project differentiates itself through a function calling interface that enables AI agent development by connecting the model to external APIs and real-time web browsing. It includes specialized capabilities for generating functional programming cod
Integrates high-resolution visual features into a shared vector space for joint visual and linguistic reasoning.
Corenet is a deep learning training framework and computer vision model library designed for developing neural networks across vision, text, and audio modalities. It functions as a distributed training orchestrator for scaling workloads across multiple compute nodes and provides a multimodal data pipeline for processing image, text, and video data. The project includes a model conversion toolkit for transforming weights and architectures between different machine learning frameworks. It also provides tools for optimizing model performance on Apple Silicon and reducing response latency in gene
Converts raw audio and image bytes into specialized tensors for neural network processing.
Genkit is an open-source framework for building AI-powered applications. It provides a unified interface for connecting to hundreds of generative AI models from multiple providers, enabling text, image, audio, and video generation through a single API. The framework structures multi-step AI interactions—including chat, retrieval-augmented generation, tool use, and agentic workflows—as composable, traceable flows with built-in streaming and state management. The framework distinguishes itself through a comprehensive developer toolkit that includes a command-line interface and a local developer
Embeds media like images into prompts for models that support multimodal input.
Gemma es una familia de modelos de lenguaje de gran tamaño de pesos abiertos basados en una arquitectura transformer solo de decodificador. Estos modelos están diseñados para la generación de texto y conversaciones multimodales, capaces de procesar y generar respuestas basadas tanto en secuencias de entrada textuales como visuales. El proyecto proporciona un modelo de IA ajustable (fine-tunable) que admite el ajuste de pesos y la adaptación de bajo rango (LoRA) para especializar el rendimiento en tareas particulares. Incluye soporte para pesos cuantizados para reducir el uso de memoria y aumentar la velocidad de inferencia en hardware limitado. La superficie de capacidades cubre la integración de IA multimodal, optimización de memoria mediante fragmentación de parámetros (sharding) y la integración de herramientas y APIs externas para recuperar datos en tiempo real. Permite además la generación de imágenes a partir de texto y el muestreo de salidas de texto estructurado.
Integrates visual and textual data by mapping different input modalities into a shared latent space for joint processing.
DeepSeek-VL2 es un modelo de lenguaje grande multimodal y sistema de visión-lenguaje diseñado para analizar escenas visuales y generar texto descriptivo. Funciona como un modelo de respuesta a preguntas visuales y fundamentación visual (visual grounding), capaz de extraer información de documentos y localizar objetos o regiones específicas dentro de imágenes basadas en descripciones textuales. El proyecto utiliza una arquitectura de mezcla de expertos (mixture-of-experts) para procesar entradas combinadas de imagen y texto. Está optimizado para la inferencia mediante prellenado incremental, lo que reduce los requisitos de memoria de GPU en el hardware. El modelo cubre el análisis de datos multimodal y la comprensión de documentos visuales, incluyendo la interpretación de gráficos y diseños. Realiza inferencia visual y fundamentación para hacer coincidir consultas textuales con el contenido visual correspondiente.
Implements cross-attention mechanisms to align visual regions with specific text tokens.
Muzic es una plataforma y framework de deep learning para el análisis, composición y síntesis de música impulsada por IA. Funciona como un framework de generación de música y herramienta de análisis, utilizando modelos de lenguaje grandes y agentes autónomos para orquestar la creación e interpretación de música simbólica y de audio. El proyecto se distingue por sus capacidades multimodales, mapeando el lenguaje natural y la música simbólica en un espacio de incrustación (embedding) conjunto compartido para clasificación zero-shot y recuperación de información. Emplea una variedad de arquitecturas especializadas, incluyendo frameworks de difusión para síntesis de audio, mecanismos de atención de grano dual para consistencia estructural de secuencias largas y un sistema híbrido que combina reglas de teoría musical con redes neuronales. La plataforma cubre una amplia gama de capacidades, incluyendo la generación de secuencias MIDI a partir de texto y letras, síntesis de voz cantada neuronal y transcripción automatizada de letras. También proporciona herramientas para el modelado de estructuras musicales, generación simbólica basada en atributos y la orquestación de herramientas musicales externas a través de agentes autónomos. Las utilidades de soporte incluyen pipelines de ingeniería de datos para la binarización de MIDI a gran escala, codificación de conjuntos de datos y procesamiento de señales de audio para la extracción de notas de melodía y alineación de voz a fonema.
Maps symbolic music and natural language into a shared joint embedding space using contrastive learning.
Este proyecto es un framework de segmentación de imágenes multimodal y un modelo de visión de texto a máscara. Sirve como un segmentador visual basado en SAM diseñado para aislar objetos distintos dentro de imágenes y videos convirtiendo prompts de lenguaje natural y otras entradas en máscaras semánticas a nivel de píxel. El sistema funciona como un framework de segmentación de imágenes multimodal que integra señales de texto, imagen y audio para generar máscaras. Incluye un rastreador de objetos de video interactivo que aísla y rastrea entidades visuales a través de fotogramas de video utilizando imágenes de referencia o consultas textuales. El framework proporciona capacidades para el etiquetado semántico de imágenes, asignando nombres de categorías a las máscaras a través de un vocabulario predefinido. También admite la edición interactiva de imágenes a través de la memoria del historial de sesiones y se refiere a la coincidencia de características para extraer objetos basados en regiones de imágenes de referencia.
Combines textual prompts, images, and audio signals into a unified embedding space to guide the segmentation process.
LightGlue is a deep learning framework designed for local feature matching and high-speed correspondence estimation between pairs of images. It functions as a computer vision matching model that identifies corresponding keypoints across different viewpoints. The system utilizes an adaptive neural network architecture that dynamically optimizes inference speed by pruning its own depth and width based on the input image pairs. This approach employs a transformer-style attention mechanism and cross-image attention to compute correlations between feature descriptors. The matching process include
Employs cross-attention mechanisms to compute correlations between feature descriptors of two different images.
EchoMimic is a multimodal human animation framework and diffusion-based video generator. It produces lifelike facial and semi-body animations of a reference image by synthesizing motion and appearance from various source data. The system enables portrait animation driven by audio, pose sequences, or driver videos. It features a landmark conditioning tool that allows for the precise control of facial movements by modifying specific landmark points. The framework covers multi-modal motion synthesis and the synchronization of reference images to match the physical movements of a target driver.
Combines audio and pose data into a unified latent space to control subject appearance and motion.