-
Immediately after restart host may switch from GPU/CUDE device to CLANG and nothing can do except shutdown other peers to find who causing the issue.
On unhealthy host issue detected by basic health …
-
Hello,
I was trying to setup local run on my mac and got this error when running: "torchserve --start --ts-config config.local.properties --foreground" And got a no module named "nvgpu" error.
…
-
### What is the issue?
Ollama indicates the model is utilizing 22GB, but nvidia-smi says it's utilizing 16GB.
The model was fully loaded and generating responses when I ran nvidia-smi.
Here's the…
-
안녕하세요. 올려주신 pdf대로 설정해보았는데, train을 실행하니 cuda out of memory가 발생합니다. 배치 사이즈는 1로 설정했습니다.
…
-
Hi thanks for the library! It seems that pytorch 2.5 is out and supports cuda 12.4. Thus it would be great if this container could be updated :)
-
### Problem Description
At the moment it seems like Rancher Desktop for Windows does not support Nvidida CUDA. I have tried both, the `containerd` and the `dockerd` engines.
Executing `nerdctl run…
-
Commands required to enable NVIDIA gpu performance counter for all user
```
echo "options nvidia NVreg_RestrictProfilingToAdminUsers=0" > /etc/modprobe.d/nvidia.conf
systemctl stop nvidia-dcgm
nvi…
-
**To Reproduce**
Steps to reproduce the behavior:
1. Use an 2XL40s machine ( gx3-48x240x2l40s in IBM Cloud )
2. Use instructlab image : registry.stage.redhat.io/rhelai1/instructlab-nvidia-rhel9:1.2-17…
-
Sometimes there is no information from sensors and nvidia-smi return "-" so consider replacing parce_line with:
```
def parse_line(line_string):
parsed_list = list(('-' if val == '-' else func(…
-
During the `ilab config init` and machine with H100 GPUs. ( a3-highgpu-8g to be specific in GCP ) will detect the H100 as being a A100.
This could raise doubts on proper identification of system.
`…