Arrakis / Models / NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 GPU requirements

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 has 32B parameters. To serve it, plan for about 76 GB of GPU memory in 16-bit, 38 GB in 8-bit or 19 GB in 4-bit, including about 20% for the KV cache and activations.

It is a mixture of experts: 128 experts, 6 of them active for each token. All of them stay in GPU memory, so plan memory for all 32B parameters.

Deploy this model Check another model

What it takes to serve it

GPU memory to run it for inference, with the smallest GPU setup that holds it and the lowest price on Arrakis right now.

Precision	Memory and the smallest setup
16-bit	76 GB: one H100 or A100, H100 from $2.25 per GPU-hour
8-bit	38 GB: one L40S, RTX 6000 Ada or RTX A6000, L40S from $0.44 per GPU-hour
4-bit	19 GB: one RTX 4090 or RTX 3090, RTX 3090 from $0.19 per GPU-hour

What it takes to fine-tune it

QLoRA trains adapters on a 4-bit frozen model, LoRA on a 16-bit one, and full fine-tuning updates every weight. Figures are before activations, which grow with batch size and sequence length.

Method	Memory and the smallest setup
QLoRA	22 GB: one RTX 4090 or RTX 3090, RTX 3090 from $0.19 per GPU-hour
LoRA	76 GB: one H100 or A100, H100 from $2.25 per GPU-hour
Full fine-tuning	505 GB: 2 × B300, from $9.20 per hour for all 2

About the model

From its card and files on Hugging Face. Memory figures here follow the GPU memory guide.

Parameters: 32B
Active for each token: 3B
Experts: 128
Experts for each token: 6
Context window (tokens): 262,144
Published in: BF16
License: other
Released: 2026-08-01
Weights: nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16

Deploy it on Arrakis

Open Models in the console, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine. Or run it yourself on any machine with enough GPU memory:

vLLM

vllm serve nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 --tensor-parallel-size 1

Deploy this model How deploying works

Questions

Live prices for every GPU are on the GPU prices page.

How much GPU memory does NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 need?

To serve it, about 76 GB in 16-bit, 38 GB in 8-bit, 19 GB in 4-bit, including about 20% for the KV cache and activations. To fine-tune it, about 22 GB with QLoRA, 76 GB with LoRA and 505 GB for full fine-tuning with Adam, before activations.

Can I run NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 on one GPU?

Yes. In 16-bit it needs about 76 GB, which fits on one H100 or A100.

What does it cost to run NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16?

In 16-bit it needs one H100 or A100; in 4-bit, one RTX 4090 or RTX 3090. Each setup on this page shows the lowest price per GPU-hour on Arrakis right now, and you pay by the hour from a prepaid balance.

How do I deploy NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 on Arrakis?

Open Models in the Arrakis console, paste nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine.

More models

Every model, and a check for any other.