Granite-Embedding-97M-Multilingual-R2 is a 97M parameter dense embedding model from the Granite Embeddings collection, designed for multilingual text embeddings. It uses a ModernBERT-based bi-encoder architecture, produces 384-dimensional vectors, and supports context lengths up to 32,768 tokens. The model is released under the Apache-2.0 license and supports 200+ languages, with enhanced support for 52 languages and programming code. It is available with ONNX and OpenVINO formats and is compatible with sentence-transformers, vLLM, and llama.cpp (GGUF).
Model size
0.10B
Format
FP16 weights
Context
32,768
Freshness 50 out of 100. Repository age: 178 days old, Apr 20, 2026. Measures age, not quality.Freshness50/100178 days old
Popularity 66 out of 100. 114,001 downloads in 30 days; 147 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.Popularity66/100114K downloads / 30d
Granite-Embedding-311M-Multilingual-R2 is a 311M-parameter dense embedding model from IBM's Granite Embeddings collection, licensed Apache-2.0. It produces 768-dimensional vectors and supports a context length of up to 32,768 tokens. The model is based on the ModernBERT architecture and supports 200+ languages, with enhanced support for 52 languages and programming code. It is released with ONNX and OpenVINO models and is tagged for sentence-transformers, feature extraction, and multilingual use.
Model size
0.31B
Format
FP16 weights
Context
32,768
Freshness 50 out of 100. Repository age: 178 days old, Apr 20, 2026. Measures age, not quality.Freshness50/100178 days old
Popularity 68 out of 100. 150,966 downloads in 30 days; 143 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.Popularity68/100151K downloads / 30d