Model Releases

IBM releases Granite Embedding Multilingual R2 open-source multilingual embedding models

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

AI summary

IBM has released two Apache 2.0 open-source multilingual embedding models: the 97M-parameter compact model (granite-embedding-97m-multilingual-r2) scores 60.3 on MTEB Multilingual Retrieval, surpassing all open sub-100M-parameter models; the 311M full-size model (granite-embedding-311m-multilingual-r2) scores 65.2, ranking second among open models under 500M and supporting Matryoshka dimensions. Both are based on the ModernBERT architecture, support 200+ languages, are trained for retrieval across 52 languages and 9 programming languages, and have a context window of 32,768 tokens. They can directly replace default models in frameworks such as sentence-transformers, LangChain, and LlamaIndex, with prebuilt ONNX and OpenVINO weights to optimize CPU inference.

Key Takeaways

  • IBM has released two Apache 2.
  • 0 open-source multilingual embedding models: the 97M-parameter compact model…
  • 3 on MTEB Multilingual Retrieval,surpassing all open sub-100M-parameter models
  • the 311M full-size model (granite-embedding-311m-multilingual-r2) scores 65.
  • 2,ranking second among open models under 500M and supporting Matryoshka dimensions.

Quick Facts

Published
Updated
Source
Hugging Face 社区博客(混合发现)
Category
Model Releases
AI score
60
Read the original Read the full article at the original source
Why it matters

97M小模型在跨语言检索上打败很多300M级选手,32K上下文让长文档处理不再鸡肋,Apache 2.0开源意味着做多语言RAG的可以直接拿来当默认嵌入模型了。

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: IBM releases Granite Embedding Multilingual R2 open-source multilingual embedding models. May 15, 2026. https://new.nat6.net/en/news/cmp5wen830hsgsljxyhi7cgwl/

View all →