[GH-ISSUE #332] Multiple GPU support #113

Closed
opened 2026-10-18 21:06:12 +00:00 by bsakaguchi · 1 comment
Owner

Originally created by @arturzangiev on GitHub (Dec 28, 2025).
Original GitHub issue: https://github.com/open-webui/helm-charts/issues/332

I have the following configuration in values.yaml, but it still uses only one GPU

ollama:
  enabled: true
  fullnameOverride: "open-webui-ollama"
  ollama:
    gpu:
      enabled: true
      type: 'nvidia'
      number: 2
    resources:
      limits:
        nvidia.com/gpu: 2
      requests:
        nvidia.com/gpu: 2
  runtimeClassName: nvidia

Do you support multiple GPUs?

Originally created by @arturzangiev on GitHub (Dec 28, 2025). Original GitHub issue: https://github.com/open-webui/helm-charts/issues/332 I have the following configuration in values.yaml, but it still uses only one GPU ``` ollama: enabled: true fullnameOverride: "open-webui-ollama" ollama: gpu: enabled: true type: 'nvidia' number: 2 resources: limits: nvidia.com/gpu: 2 requests: nvidia.com/gpu: 2 runtimeClassName: nvidia ``` Do you support multiple GPUs?
Author
Owner

@james-westbrook commented on GitHub (Apr 8, 2026):

Hi @arturzangiev, our chart does direct passthrough to the Ollama subchart, so anything their chart supports, you can access through the subchart. I believe you have an issue with your posted values. You could try this instead:

ollama:
  enabled: true
  fullnameOverride: "open-webui-ollama"
  ollama:
    gpu:
      enabled: true
      type: 'nvidia'
      number: 2       # requests 2 GPUs — automatically sets resource limits
  runtimeClassName: nvidia
  persistentVolume:
    enabled: true
<!-- gh-comment-id:4210297756 --> @james-westbrook commented on GitHub (Apr 8, 2026): Hi @arturzangiev, our chart does direct passthrough to the Ollama subchart, so anything their chart supports, you can access through the subchart. I believe you have an issue with your posted values. You could try this instead: ``` ollama: enabled: true fullnameOverride: "open-webui-ollama" ollama: gpu: enabled: true type: 'nvidia' number: 2 # requests 2 GPUs — automatically sets resource limits runtimeClassName: nvidia persistentVolume: enabled: true ```
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
open-webui/helm-charts#113
No description provided.