Skip to content
AETRONConstructor
My neuronetsRegistrySpacesDocsFAQ
Sign in
AETRON

A constructor for AI services on AETRON’s decentralised network of GPU miners.

ProductCreate a neuronetModel catalogueDatasetsSpacesMy payouts
ResourcesDocumentationAPITimings & statusesFAQ
Networkaetron.ioPrivacy & verification
© 2026 AETRONTestnet
Models/nemotron-3-super-120b
N

Nemotron-3 Super 120B-A12B

NVIDIA · Nemotron-3
Create a neuronet with this model →
LLMNVIDIA-Open-ModelpublicTrusted120B params
moe

Super 120B-A12B: single 80 GB card at 4-bit.

Super 120B-A12B: single 80 GB card at 4-bit.

At a glance

  • Task: LLM (InferenceV1)
  • Params: 120B
  • License: NVIDIA-Open-Model (commercial use OK)
  • Hardware: 272 GB+ VRAM
  • dtype: Bf16, Fp16
Create a neuronet with this model →
Overview
Task
LLM · InferenceV1
Family
Nemotron-3
Tier
Trusted
Status
active
Runtime spec
dtype
Bf16
attention
FlashAttn2
compile
ReduceOverhead
seed
0
gpu_arch
Sm90
Runs on
Datacenter
Hardware
Min VRAM
272 GB
Recommended
307 GB
Tensor parallel
≥2
TEE archs
Sm90
Verification defaults
Tier
B
Trust
Level1
Security
Standard
Privacy
P0
Min pulse score
100
License
spdx
NVIDIA-Open-Model ↗
Commercial use
✓
Redistribution
✓
Source & provenance
Repo
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 ↗
Revision
d51eab0d1f
Added
2026-06-08
Verified
2026-06-08