Tag

#embeddings

Every item tagged embeddings, newest first.

5 items

Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context — Best Sub-100M Retrieval Quality

IBM released Granite Embedding Multilingual R2 under Apache 2.0, offering 32K context and sub-100M retrieval quality. This open multilingual embedding model supports 100+ languages and targets builders seeking high-quality, locally deployable models for semantic search and retrieval tasks. The model's performance is competitive with larger, closed alternatives.

Key takeaways

Released under Apache 2.0 for open use.
32K context window for longer input sequences.
Competitive retrieval quality below 100M parameters.

HHugging Face Blog#open-source #multilingual #embeddings #retrieval

modelsApr 9

Multimodal Embedding & Reranker Models with Sentence Transformers

Hugging Face released multimodal embedding and reranker models using Sentence Transformers, enabling joint text and image encoding for applications like image search and visual question answering. These models allow you to build multimodal applications with a single, unified embedding space. The Sentence Transformers library provides a simple interface for using these models.

Key takeaways

Multimodal models encode text and images in a single space.
Enables applications like image search and visual question answering.
Sentence Transformers library provides a simple interface.

HHugging Face Blog#multimodal #sentence-transformers #embeddings

modelsSep 4

Welcome EmbeddingGemma, Google's new efficient embedding model

Google released EmbeddingGemma, a lightweight embedding model targeting efficient, high-performance vector search and retrieval applications. EmbeddingGemma is optimized for low-latency, low-memory use cases, making it suitable for edge deployments and mobile devices. You can use EmbeddingGemma for applications like semantic search, recommendation systems, and natural language processing tasks. The model is available on Hugging Face for download and integration.

Key takeaways

EmbeddingGemma is optimized for low-latency, low-memory use cases.
The model is suitable for edge deployments and mobile devices.
EmbeddingGemma is available on Hugging Face for download.

HHugging Face Blog#efficient-ai #embeddings #edge-ai

toolsMar 15

CPU Optimized Embeddings with 🤗 Optimum Intel and fastRAG

Hugging Face released CPU optimized embeddings with Intel and fastRAG, allowing for faster and more efficient processing of large datasets. This optimization targets builders who need to improve the performance of their models on CPU infrastructure. The collaboration between Hugging Face and Intel aims to provide better support for CPU-based deployments. This development can help reduce the costs associated with running large language models.

Key takeaways

CPU optimized embeddings available for faster processing
Collaboration between Hugging Face and Intel for better CPU support
Improved performance for large language models on CPU infrastructure

HHugging Face Blog#cpu-optimization #embeddings #collaboration

tutorialsJun 23

Getting Started With Embeddings

Hugging Face published a guide on getting started with embeddings, covering the basics and applications of embeddings in natural language processing. The guide is intended for developers and researchers who want to learn about embeddings and how to use them in their projects. Embeddings are a fundamental concept in NLP and are used in many state-of-the-art models. By understanding embeddings, you can improve your NLP models and applications.

Key takeaways

Embeddings are a fundamental concept in NLP.
Hugging Face provides a guide for getting started with embeddings.
Embeddings are used in many state-of-the-art NLP models.

HHugging Face Blog#nlp #embeddings #tutorials