33.7B
- input
- $0.00
- output
- $0.00
- Context
- 33Ktokens
- Listed
- 3mo ago
- Size
- 1Bparameters
Cost/1M tokens
Nemotron 3 Embed 1B is an open text embedding model released by NVIDIA in the Nemotron 3 Embed line, derived from Mistral's Ministral-3-3B-Instruct-2512 via structured pruning and distillation. Per its model card, it is optimized for retrieval and semantic similarity tasks and was evaluated across 34 languages for multilingual and cross-lingual retrieval. It is built to serve as a component in text-based Retrieval-Augmented Generation, including agentic RAG workflows.
How it compares
Size class
Tiny1BAn open-weight model under 5 billion parameters, small enough for phones, laptops and edge devices.
Open models of a similar size from other labs
Vs. your stack
Add the models you use to see how this one compares.
Sign inFamily tree
Models trained from one another, as their makers describe it.
Ministral 3 3BMistral AI · 2025
Frequently asked questions
Running locally
About 1.2 GB at Q4, 1.7 GB at Q8 and 2.8 GB at FP16. These are estimates from the 1B parameter count, covering weights plus about 15% and 0.5 GB for a short context.
Yes. It fits at FP16 (estimated 2.8 GB) on a 24 GB, 16 GB, 12 GB or 8 GB GPU.
Yes. At Q4 it needs an estimated 1.2 GB. Assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 8 GB.
Privacy and data
Yes. Nemotron 3 Embed 1B has open weights, so you can host it yourself for full control of your data.
Capabilities and specs
The context window is 33K tokens (32,768) and the maximum output is 29K tokens (29,491).
No. It only reads text.
Quality and benchmarks
LFM2.5-Embedding-350M by Liquid AI, voyage-4-large by VoyageAI by MongoDB at a blended price of $0.09 per 1M tokens, and voyage-4 by VoyageAI by MongoDB at a blended price of $0.045 per 1M tokens. They are matched on model kind and release date, with no comparable price.
LFM2.5-Embedding-350MNemotron 3 Embed 1B vs LFM2.5-Embedding-350Mvoyage-4-largeNemotron 3 Embed 1B vs voyage-4-largevoyage-4Nemotron 3 Embed 1B vs voyage-4
Versions and lineage
Nemotron 3 Embed 1B has a 33K token (32,768) context versus 128K tokens (128,000) for Nemotron 3 Content Safety. Embed 1B costs $0.00 input and $0.00 output per 1M tokens and was released July 16, 2026, while Content Safety was released March 16, 2026.
Nemotron 3 Content SafetyNemotron 3 Embed 1B vs Nemotron 3 Content Safety
Nemotron 3 Embed 1B is fine-tuned from Ministral 3 3B.
Licensing and access
Open weights. Nemotron 3 Embed 1B is released with open weights.
From the Hugging Face repository nvidia/Nemotron-3-Embed-1B-BF16.
Basics
NVIDIA. It was first listed on OpenRouter on July 16, 2026.
Yes. It has open weights.
Generated from Vertical Tabs data, updated November 10, 2026.
More in Nemotron
33.7B1BNemotron 3.5 Embed 1B
8BNemotron 3.5 Embed 8B
8BNemotron 3.5 Rerank 8B
Nemotron 3.5 Super VL
120BNemotron 3.5 Super 120B A12B
0.6BNemotron 3.5 ASR Streaming Multilingual 0.6B
30BNemotron 3.5 Lightning
550BNemotron 3 Ultra
4.3BNemotron 3.5 Content Safety
30BNemotron 3 Nano Omni
30B
Discussion
No comments yet
Be the first to share your thoughts.