OPEN MODELS / Vision language models
Run NVIDIA-Nemotron-Nano-12B-v2-VL-FP8 on a cloud GPU
NVIDIA-Nemotron-Nano-12B-v2-VL-FP8 is a Vision language models model by nvidia. See how much GPU memory it needs, which GPUs it fits on today and what it costs to run it yourself, billed per second.
GPUs it fits on today
Our estimate is about 17 GB for inference with F8_E4M3 weights, including working memory. Real use depends on resolution, batch size and context length. With 4-bit quantization it needs about 11 GB.
| GPU | Memory | Per hour | 10 hours | |
|---|---|---|---|---|
| RTX 2080 Ti | 22 GB | $0.25 | $2.50 | Rent |
| Tesla P40 | 24 GB | $0.27 | $2.70 | Rent |
| RTX 4000Ada | 20 GB | $0.39 | $3.90 | Rent |
| RTX 3090 | 24 GB | $0.45 | $4.50 | Rent |
| RTX 3090 Ti | 24 GB | $0.45 | $4.50 | Rent |
Live starting prices from our catalog. You pay per second while the machine runs.
How to run NVIDIA-Nemotron-Nano-12B-v2-VL-FP8
- 01
Rent a GPU
Pick one of the GPUs above. Billing is per second; stop the machine and you stop paying.
- 02
Open a terminal or Jupyter
Connect over SSH or open Jupyter in your browser. PyTorch and CUDA are ready to use.
- 03
Download and run the model
Run the commands below. The weights download straight from Hugging Face to your machine.
pip install -U vllm
vllm serve nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-FP8 --host 0.0.0.0 --port 8000Why run it yourself?
Run document and image understanding on your own GPU: no per-image fee, and files stay on your machine.
Questions
- How much GPU memory does NVIDIA-Nemotron-Nano-12B-v2-VL-FP8 need?
- By our estimate about 17 GB for inference (13B parameters). Higher resolution, longer context or bigger batches need more.
- How much does it cost to run NVIDIA-Nemotron-Nano-12B-v2-VL-FP8?
- From $0.25 per hour on RTX 2080 Ti, billed per second. Ten hours cost about $2.50.
- Do I need a subscription?
- No. You pay only for the time your machine runs, from your USD balance. Stop it whenever you are done.
Try it on your own GPU in minutes
Pick a GPU, open Jupyter or SSH and run any open model. You pay per second and can stop anytime; no subscription, no setup fee.
- Per-second billing in USD
- PyTorch and CUDA ready in minutes
- Rent from ChatGPT or Claude too