1× RTX 507012 GB VRAM
Nosana network, hosts worldwide
Details
Stock and price checked 10 min ago.
Self-service: you start it yourself, with no order or sales contact.
nvidia-5070
Full price
€0.13
per hour, whole instance, 1 GPU
datalab-to
chandra-ocr-2
image text to textopenrail1.4M downloadsHugging Face
Cheapest GPU in stock that fits
A single GPU with at least 12 GB for inference.
View offer
GPU memory it needs (estimate)
Rule of thumb from the model's size. Real needs depend on batch size, context length and framework.
Inference (full precision)
12 GB
Inference (8-bit)
5.9 GB
Inference (4-bit)
3.0 GB
LoRA fine-tuning
13 GB
QLoRA (8-bit base)
6.4 GB
Full fine-tuning
47 GB