Text generation · served with ONNX Runtime GenAI · ↓ 0 downloads a month · View on Hugging Face
Prince-1/Nemotron-Terminal-14B-Onnx needs about 36.8 GB of GPU memory. It runs on one Quadro RTX 8000 from $0.31/h right now; best value: RTX PRO 5000 at $0.68/h.
Weights
30 GB
GPU memory needed
about 36.8 GB
Memory · disk
48.3 GB · 93 GB
Licence
other
Hardware and deploy
Deploying ONNX models from the catalog is coming soon. Today, download the files from Hugging Face and upload them in Inference.
Questions
How much GPU memory does Nemotron-Terminal-14B-Onnx need?
About 36.8 GB: 29.56 GB of weights and the runtime's own memory.
What is the cheapest GPU to run Nemotron-Terminal-14B-Onnx?
Right now one Quadro RTX 8000 at $0.31 an hour; RTX PRO 5000 at $0.68 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.
Can Nemotron-Terminal-14B-Onnx run on an RTX 4090?
Not on one: it needs about 36.8 GB and an RTX 4090 has 24 GB. Use a bigger GPU, or a smaller (quantized) version of the model.
Prices are per hour for the whole machine and its disk, right now. Speed is Vast's DLPerf score of the machine, relative to the cheapest. Best value: the most speed per dollar. Only machines with a Vast reliability above 95% that download at 400 Mbit/s or more are offered: the machine fetches the model itself.