llama-embed-nemotron-8b (nvidia) specs, prices, where it runs and its benchmark standing

Name: llama-embed-nemotron-8b (nvidia)
Author: Nvidia

open weights · top 10 on 1 of 1 · released Oct 23, 2025

Provider: Nvidia
Release date: Oct 23, 2025; Boards
Parameters: 7.5B total; Hugging Face
Modalities: Text to Embeddings; Hugging Face
License: Customized NSCL V1; Hugging Face
Hosting: self-host: weights on Hugging Face
Hardware: Estimate: about 15 GB of weights as published (BF16), about 7.5 GB at 8-bit, about 3.8 GB at 4-bit, plus memory for the context; Hugging Face config
Weights: Open weights
Benchmarks: 1 board captured

Rank	Board	Category	Score	Of	Over time
3	MTEBEmbeddings · as “nvidia/llama-embed-nemotron-8b”	Embeddings	69.5%	96

Category ranks, more models and catalog changes

Facts