Model3d ago
Bing Releases Harrier Embedding Model
Harrier tops the multilingual MTEB v2 benchmark, scoring 1,200 points across 100+ languages. Microsoft released the model under an open‑source license, enabling developers to embed text in any supported language without costly APIs. The release gives practitioners a ready‑to‑use, high‑performance embedding engine for multilingual NLP pipelines. It also includes pre‑trained weights and a simple Python API.