Embeddings (features) · served with ONNX Runtime · ↓ 6.8K downloads a month · View on Hugging Face
perplexity-ai/pplx-embed-context-v1-4b needs about 25.2 GB of GPU memory. It runs on one Nvidia V100 from $0.17/h right now; best value: RTX 5090 at $0.52/h.
Weights
16 GB
GPU memory needed
about 25.2 GB
Memory · disk
36.2 GB · 58 GB
Licence
mit
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 pplx-embed-context-v1-4b need?
About 25.2 GB: 16.11 GB of weights and the runtime's own memory.
What is the cheapest GPU to run pplx-embed-context-v1-4b?
Right now one Nvidia V100 at $0.17 an hour; RTX 5090 at $0.52 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.
Can pplx-embed-context-v1-4b run on an RTX 4090?
Not on one: it needs about 25.2 GB and an RTX 4090 has 24 GB. An RTX 5090 (32 GB) is enough.
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.