github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #2395] Multi-GPU setup of Tesla P100s is slow #1393

New Issue
Closed
opened 2026-04-12 11:13:25 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 11:13:25 -05:00
Owner
Copy Link

Originally created by @PhilipAmadasun on GitHub (Feb 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2395

Originally assigned to: @dhiltgen on GitHub.

A multi-GPU setup of Tesla P100s is very slow compared to a single RTX 4090. I am using the 0.1.22 version of ollama. Is there something wrong with the Teslas? Are they just bad GPUs? I was told to try to run ollama on just one of them to see what happens, if that might indeed make ollama run faster I am not sure how to go about it. Is there some way to disable GPU so ollama only runs on a single Tesla GPU?

Originally created by @PhilipAmadasun on GitHub (Feb 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2395 Originally assigned to: @dhiltgen on GitHub. A multi-GPU setup of Tesla P100s is very slow compared to a single RTX 4090. I am using the 0.1.22 version of ollama. Is there something wrong with the Teslas? Are they just bad GPUs? I was told to try to run ollama on just one of them to see what happens, if that might indeed make ollama run faster I am not sure how to go about it. Is there some way to disable GPU so ollama only runs on a single Tesla GPU?
GiteaMirror added the gpu label 2026-04-12 11:13:25 -05:00
GiteaMirror commented 2026-04-12 11:13:27 -05:00
Author
Owner
Copy Link

@opyate commented on GitHub (Mar 5, 2024):

Is there some way to disable GPU so ollama only runs on a single Tesla GPU?

You might be able to get the GPU's ID from nvidia-smi, and then just use that ID to isolate the GPU for ollama, e.g.

CUDA_VISIBLE_DEVICES=1 ollama serve
<!-- gh-comment-id:1978680686 --> @opyate commented on GitHub (Mar 5, 2024): > Is there some way to disable GPU so ollama only runs on a single Tesla GPU? You might be able to get the GPU's ID from `nvidia-smi`, and then just use that ID to isolate the GPU for ollama, e.g. ``` CUDA_VISIBLE_DEVICES=1 ollama serve ```
GiteaMirror commented 2026-04-12 11:13:28 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Mar 21, 2024):

If you're still seeing unexpected performance behavior, please share your server log so we can analyze what's going on.

<!-- gh-comment-id:2012375832 --> @dhiltgen commented on GitHub (Mar 21, 2024): If you're still seeing unexpected performance behavior, please share your server log so we can analyze what's going on.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label gpu
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#1393
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?