LLM EXPLORER 56,604 MODELS INDEXED

NVIDIA Nemotron Nano 12B V2 by nvidia

By nvidia · 199598 downloads
HF Repo ↗ Copy URL

NVIDIA Nemotron Nano 12B V2 is an open-source language model by nvidia. Features: 12b LLM, VRAM: 24.6GB, Context: 128K, License: other, LLM Explorer Score: 0.3.

  Arxiv:2412.02595   Arxiv:2504.03624   Arxiv:2508.14444 Base model:finetune:nvidia/nvi... Base model:nvidia/nvidia-nemot...   Conversational Dataset:nvidia/nemotron-cc-mat...   Dataset:nvidia/nemotron-cc-v2 Dataset:nvidia/nemotron-post-t... Dataset:nvidia/nemotron-post-t... Dataset:nvidia/nemotron-pretra... Dataset:nvidia/nemotron-pretra...   De   Deploy:azure   En   Endpoints compatible   Es   Fr   It   Ja   Nvidia   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2 

NVIDIA Nemotron Nano 12B V2 Parameters and Internals

LLM NameNVIDIA Nemotron Nano 12B V2
Repository 🤗https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2 
Base Model(s)  ...IDIA Nemotron Nano 12B V2 Base   nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base
Model Size12b
Required VRAM24.6 GB
Updated2026-07-11
Maintainernvidia
Model Typenemotron_h
Model Files  4.8 GB: 1-of-6   4.9 GB: 2-of-6   4.8 GB: 3-of-6   4.9 GB: 4-of-6   3.9 GB: 5-of-6   1.3 GB: 6-of-6
Supported Languagesen es fr de it ja
Model ArchitectureNemotronHForCausalLM
Licenseother
Context Length131072
Model Max Length131072
Transformers Version4.51.3
Tokenizer ClassPreTrainedTokenizer
Vocabulary Size131072
Torch Data Typebfloat16

Best Alternatives to NVIDIA Nemotron Nano 12B V2

Best Alternatives
Context / RAM
Downloads
Likes
NVIDIA Nemotron Nano 9B V2128K / 17.7 GB301038506
...IDIA Nemotron Nano 12B V2 Base128K / 24.6 GB360691
Note: green Score (e.g. "73.2") means that the model is better than nvidia/NVIDIA-Nemotron-Nano-12B-v2.