Nemotron-3-Embed-1B-NVFP4

Image Generation

nvidia · nvidia · v3 · released 2026-01-01

About

wer

Architecture

Type: encoder
Parameters: 1.0B

wer

Memory

Weights (BF16): 2.00 GB
Activation estimate: 0.00 GB

Pricing

Free — open weights

Self-host on your own GPU. The calculator surfaces GPU-hours cost on the hardware page instead of an API price.

Provenance

Hugging Face: nvidia/Nemotron-3-Embed-1B-NVFP4

vllmsafetensorsministral3texttext-embeddingsretrievalsemantic-searchnvfp4