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

Published: 2026-05-14 · Updated: 2026-05-14 · Category: Open Source

Source page: https://www.sipply.dev/story/granite-embedding-multilingual-r2-open-apache-2-0-bb42b79b · 中文: https://www.sipply.dev/zh/story/granite-embedding-multilingual-r2-open-apache-2-0-bb42b79b

## Summary

Hugging Face has introduced two new multilingual embedding models, granite-embedding-311m-multilingual-r2 and granite-embedding-97m-multilingual-r2, both released under the Apache 2.0 license. These models are designed to support over 200 languages and can handle a context of up to 32,000 tokens, which significantly enhances their usability in multilingual applications. The models have been noted for achieving exceptional retrieval quality, particularly for datasets under 100 million entries, positioning them as a valuable resource for developers and researchers working in the field of natural language processing.

## Key takeaways

- Hugging Face's new models support a vast array of languages and large context sizes.
- The models are open-source under Apache 2.0, promoting accessibility and collaboration.
- They provide top-tier retrieval quality for multilingual applications, especially with smaller datasets.

## Why it matters

The release of these models represents a significant advancement in multilingual NLP capabilities, enabling more effective communication and information retrieval across diverse languages. This can lead to improved user experiences in global applications.

- **Product managers:** For product managers, these models offer an opportunity to enhance product features by integrating advanced multilingual capabilities, potentially increasing market reach and user engagement.
- **Developers:** Developers can leverage these models to build more robust multilingual applications, improving the quality of language processing tasks and expanding the functionality of their software solutions.
- **Students and job seekers:** Students and job-seekers can benefit from familiarity with these cutting-edge models, as they represent a growing area of demand in the job market, particularly in roles focused on AI and machine learning.

## Sources

- [Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context — Best Sub-100M Retrieval Quality](https://huggingface.co/blog/ibm-granite/granite-embedding-multilingual-r2) (Hugging Face Blog, 2026-05-14)

## 中文

### 花岗岩嵌入多语言 R2：开放的 Apache 2.0 多语言嵌入，支持 32K 上下文 — 最佳的 Sub-100M 检索质量

Hugging Face 推出了两个新的多语言嵌入模型，granite-embedding-311m-multilingual-r2 和 granite-embedding-97m-multilingual-r2，均在 Apache 2.0 许可下发布。这些模型旨在支持超过 200 种语言，并能够处理多达 32,000 个标记的上下文，显著增强了它们在多语言应用中的可用性。这些模型在检索质量方面表现出色，特别是在数据集少于 1 亿条目的情况下，使其成为自然语言处理领域开发者和研究人员的宝贵资源。

**要点：**

- Hugging Face 的新模型支持大量语言和大上下文大小。
- 这些模型在 Apache 2.0 下开源，促进了可访问性和协作。
- 它们为多语言应用提供顶级检索质量，特别是在较小的数据集上。

**为什么重要：** 这些模型的发布代表了多语言 NLP 能力的重大进展，使得在不同语言之间的沟通和信息检索更加有效。这可以改善全球应用中的用户体验。

---
Sipply: Sip AI News Simply. https://www.sipply.dev