LLM EXPLORER 60,587 MODELS INDEXED

NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning by unsloth

By unsloth · 502 downloads
HF Repo ↗ Copy URL

NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning is an open-source language model by unsloth. Features: 30b LLM, VRAM: 66.1GB, License: other, LLM Explorer Score: 0.26.

Base model:finetune:nvidia/nem... Base model:nvidia/nemotron-3-n...   Conversational   Custom code Dataset:nvidia/nemotron-post-t... Dataset:nvidia/nemotron-pre-tr...   De   En   Es   Fr   It   Ja   Multimodal   Nemotron-3 Nemotronh nano omni reasoning ...   Nvidia   Region:us   Safetensors   Sharded   Tensorflow   Unsloth   Zh
Model Card on HF 🤗: https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning 

NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning Parameters and Internals

LLM NameNVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning
Repository 🤗https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning 
Base Model(s)  ...no Omni 30B A3B Reasoning BF16   nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16
Model Size30b
Required VRAM66.1 GB
Updated2026-06-25
Maintainerunsloth
Model TypeNemotronH_Nano_Omni_Reasoning_V3
Model Files  4.0 GB: 1-of-17   4.0 GB: 2-of-17   4.0 GB: 3-of-17   4.0 GB: 4-of-17   4.0 GB: 5-of-17   4.0 GB: 6-of-17   4.0 GB: 7-of-17   4.0 GB: 8-of-17   4.0 GB: 9-of-17   4.0 GB: 10-of-17   4.0 GB: 11-of-17   4.0 GB: 12-of-17   4.0 GB: 13-of-17   4.0 GB: 14-of-17   4.0 GB: 15-of-17   4.0 GB: 16-of-17   2.1 GB: 17-of-17
Supported Languagesen fr es it de ja zh
Model ArchitectureNemotronH_Nano_Omni_Reasoning_V3
Licenseother
Model Max Length262144
Transformers Version4.57.6
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<SPECIAL_999>
Torch Data Typebfloat16

Best Alternatives to NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning

Best Alternatives
Context / RAM
Downloads
Likes
...no Omni 30B A3B Reasoning BF160K / 66.1 GB447327407
...A3B Reasoning Abliterated BF160K / 66.1 GB170
...no Omni 30B A3B Reasoning Bf160K / 65.8 GB1261
...o Omni 30B A3B Reasoning Mxfp80K / 34.8 GB1241
...no Omni 30B A3B Reasoning 4bit0K / 19.6 GB3322
...no Omni 30B A3B Reasoning 8bit0K / 35.7 GB1651
Note: green Score (e.g. "73.2") means that the model is better than unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning.