Nemotron-3-Embed-1B-BF16

NVIDIA🇺🇸 United States
active
Context window33K tokens

Version History

1Bmajor

Initial release of 1.14B parameter multilingual embedding model derived from Ministral-3B through two rounds of structured pruning and distillation.

Coverage

model releaseNVIDIA

NVIDIA Releases Nemotron-3-Embed-1B-BF16: 1.14B Parameter Multilingual Embedding Model with 2048-Dimensional Vectors

NVIDIA has released Nemotron-3-Embed-1B-BF16, a 1.14 billion parameter text embedding model supporting 34 languages with a 32,768 token context window. The model generates 2048-dimensional embeddings and was derived from Ministral-3-3B-Instruct-2512 through two rounds of structured pruning and distillation, first to 2B then to 1.14B parameters.

2 min read