No GPU Available
Hello, I have using the 'runpodctl' workflow with dev and deploy commands and I have one network volume attached in EU-RO-1 zone. Currently whenever I run 'runpodctl project dev', CLI says that there are no GPUs available. Is it true or actually no GPU is left? Here is the part related inside my 'runpod.toml' file:
base_image = "runpod/base:0.6.1-cuda12.2.0"
gpu_types = [
"NVIDIA GeForce RTX 4080", # 16GB
"NVIDIA RTX A4000", # 16GB
"NVIDIA RTX A4500", # 20GB
"NVIDIA RTX A5000", # 24GB
"NVIDIA GeForce RTX 3090", # 24GB
"NVIDIA GeForce RTX 4090", # 24GB
"NVIDIA RTX A6000", # 48GB
"NVIDIA A100 80GB PCIe", # 80GB
]
gpu_count = 1
volume_mount_path = "/runpod-volume"
ports = "4040/http, 7270/http, 22/tcp" # FileBrowser, FastAPI, SSH
container_disk_size_gb = 100
3 Replies
4090, A4500, and 4000 Ada are available, so not sure why it says none are available.
I see 4000 Ada isn't in your list though.
I have added more but it is same
I am using 'runpodctl' version 1.14.2. Is something changed related to gpu type definitions recently?