Granite Embedding Releases New Multilingual R2 Model
Granite Embedding has released a new multilingual embedding model, R2, which is smaller at 97M parameters but outperforms other sub-100M models on MTEB Multilingual Retrieval.
01
Evidence
- HHugging Face BlogCompany2026-05-15 02:55
> Enterprise-Ready by Design A Strong Sub-100M Multilingual Model What Changed from R1 Training the Full-Size 311M Model Building the compact 97M Multilingual model Benchmark Results Multilingual Retrieval Speed and Throughput Matryoshka Embeddings (311M) Cross-lingual Retrieval…
- HHugging Face BlogCompany2026-05-15 02:55
Try The Models TL;DR: Two new Apache 2.0 multilingual embedding models built on ModernBERT — a 97M-parameter compact model that beats every open sub-100M multilingual embedder on MTEB Multilingual Retrieval (60.3), and a 311M full-size model that scores 65.2 on MTEB Multilingual…
- HHugging Face BlogCompany2026-05-15 02:55
Try The Models TL;DR: Two new Apache 2.0 multilingual embedding models built on ModernBERT — a 97M-parameter compact model that beats every open sub-100M multilingual embedder on MTEB Multilingual Retrieval (60.3), and a 311M full-size model that scores 65.2 on MTEB Multilingual…