Flow - 4x faster than typing
Mini toolsNemotron 3 Embed 8B BF16
By NVIDIA
Model family: Nemotron
Nemotron-3-Embed-8B-BF16 is an 8B parameter multilingual text embedding model from NVIDIA, built on a Ministral-3-8B based encoder with a 4096 dimensional output space and support for dynamic embedding sizes via truncation. It processes text sequences up to 32768 tokens and produces dense vector embeddings suited for retrieval augmented generation, semantic similarity search, code search, and agent memory systems. Trained and evaluated across 34 languages, it targets multilingual and cross-lingual retrieval tasks such as building question and answer applications over large text corpora. It is the flagship, highest accuracy model in the Nemotron 3 Embed collection, ranking first overall on the RTEB benchmark, and is best suited for precision critical retrieval and high stakes enterprise RAG deployments.
New
Text
Overview
Nemotron-3-Embed-8B-BF16 is NVIDIA's flagship multilingual text embedding model, built on a Ministral-3-8B based encoder. It generates dense vector representations for retrieval, semantic search, and RAG across 34 languages, and ranks first overall on the RTEB benchmark for embedding quality and agentic retrieval performance.
About NVIDIA
Industry:
Computer Hardware Manufacturing
Company Size:
42000
Location:
Santa Clara, California, US
Website:
nvidia.com
Last updated: July 17, 2026
