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 #904] centos8 use gpu #26199

New Issue
Closed
opened 2026-04-22 02:15:27 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-04-22 02:15:27 -05:00
Owner
Copy Link

Originally created by @tuoxin126 on GitHub (Oct 25, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/904

How to make my Centos8 system call GPU. Currently, Ollama can be used, but it calls the CPU. Running olama run llama2 chinese: 13b chat fp16 can directly reply. There is no process for selecting a GPU. I couldn't find the corresponding parameter by looking at the help.
From langchain.llms import Ollama
Llm=Ollama (model="llama2 chinese: 13b chat fp16")
While True:
#Receive user input
Input_ Text=input ("Please enter text (enter 'exit' to exit):")
#Check if you want to exit the loop
If input_ Text. lower()=='exit ':
Break
#Using the predict() method to generate text
Res=llm. predict (input_text)
#Print generated text
Print (res)
Using Python code for testing without selecting the location of the GPU, who knows how to call the GPU
. thanks

Originally created by @tuoxin126 on GitHub (Oct 25, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/904 How to make my Centos8 system call GPU. Currently, Ollama can be used, but it calls the CPU. Running olama run llama2 chinese: 13b chat fp16 can directly reply. There is no process for selecting a GPU. I couldn't find the corresponding parameter by looking at the help. From langchain.llms import Ollama Llm=Ollama (model="llama2 chinese: 13b chat fp16") While True: #Receive user input Input_ Text=input ("Please enter text (enter 'exit' to exit):") #Check if you want to exit the loop If input_ Text. lower()=='exit ': Break #Using the predict() method to generate text Res=llm. predict (input_text) #Print generated text Print (res) Using Python code for testing without selecting the location of the GPU, who knows how to call the GPU . thanks
GiteaMirror added the bug label 2026-04-22 02:15:27 -05:00
GiteaMirror commented 2026-04-22 02:15:28 -05:00
Author
Owner
Copy Link

@mchiang0610 commented on GitHub (Oct 25, 2023):

@tuoxin126 may I ask what are your system specs to help us look into this?

<!-- gh-comment-id:1779795393 --> @mchiang0610 commented on GitHub (Oct 25, 2023): @tuoxin126 may I ask what are your system specs to help us look into this?
GiteaMirror commented 2026-04-22 02:15:28 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 26, 2023):

@tuoxin126 may I ask what are your system specs to help us look into this?

CentOS Linux release 8.5.2111
Cuda compilation tools, release 11.7, V11.7.64
ollama version 0.1.3

Other algorithm models can call GPU

<!-- gh-comment-id:1780275136 --> @tuoxin126 commented on GitHub (Oct 26, 2023): > @tuoxin126 may I ask what are your system specs to help us look into this? CentOS Linux release 8.5.2111 Cuda compilation tools, release 11.7, V11.7.64 ollama version 0.1.3 Other algorithm models can call GPU
GiteaMirror commented 2026-04-22 02:15:29 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 26, 2023):

Ollama

Is it necessary to use Ollama in Docker to use GPU?
No matter how configured, it cannot run on GPU directly under Linux, right？
I hope to dispel doubts. Hard work

<!-- gh-comment-id:1780535207 --> @tuoxin126 commented on GitHub (Oct 26, 2023): > Ollama Is it necessary to use Ollama in Docker to use GPU? No matter how configured, it cannot run on GPU directly under Linux, right？ I hope to dispel doubts. Hard work
GiteaMirror commented 2026-04-22 02:15:29 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Oct 27, 2023):

No requirement for docker on Linux. Some just choose to use docker. What gpu do you have?

<!-- gh-comment-id:1782296371 --> @technovangelist commented on GitHub (Oct 27, 2023): No requirement for docker on Linux. Some just choose to use docker. What gpu do you have?
GiteaMirror commented 2026-04-22 02:15:30 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 27, 2023):

No requirement for docker on Linux. Some just choose to use docker. What gpu do you have?

NVIDIA A100-SXM

<!-- gh-comment-id:1782306912 --> @tuoxin126 commented on GitHub (Oct 27, 2023): > No requirement for docker on Linux. Some just choose to use docker. What gpu do you have? NVIDIA A100-SXM
GiteaMirror commented 2026-04-22 02:15:30 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 27, 2023):

No requirement for docker on Linux. Some just choose to use docker. What gpu do you have?

No requirement for docker on Linux. Some just choose to use docker. What gpu do you have?

Thank you for your clarification

<!-- gh-comment-id:1782307243 --> @tuoxin126 commented on GitHub (Oct 27, 2023): > No requirement for docker on Linux. Some just choose to use docker. What gpu do you have? > No requirement for docker on Linux. Some just choose to use docker. What gpu do you have? Thank you for your clarification
GiteaMirror commented 2026-04-22 02:15:30 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Oct 30, 2023):

Hi @tuoxin126 . Ollama should use the GPU by default. Here's a few things you can try:

  • Make sure nvidia drivers are installed. Running nvidia-smi should work.
  • Try running ollama run <model>. Does this use the GPU?

The Linux installer script will install Nvidia drivers if they aren't installed yet:

curl https://ollama.ai/install.sh | sh

If not it may be something we can look into further to help debug 😊 . I'll close this issue for now but please re-open if above doesn't help

<!-- gh-comment-id:1786120719 --> @jmorganca commented on GitHub (Oct 30, 2023): Hi @tuoxin126 . Ollama should use the GPU by default. Here's a few things you can try: * Make sure nvidia drivers are installed. Running `nvidia-smi` should work. * Try running `ollama run <model>`. Does this use the GPU? The Linux installer script will install Nvidia drivers if they aren't installed yet: ``` curl https://ollama.ai/install.sh | sh ``` If not it may be something we can look into further to help debug 😊 . I'll close this issue for now but please re-open if above doesn't help
GiteaMirror commented 2026-04-22 02:15:31 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 31, 2023):

  • ollama run

Sorry for my question. In Ollama run mode, GPU can be used, so what is the problem with my Python code. It does not use a GPU.

<!-- gh-comment-id:1786281779 --> @tuoxin126 commented on GitHub (Oct 31, 2023): > * ollama run Sorry for my question. In Ollama run mode, GPU can be used, so what is the problem with my Python code. It does not use a GPU.
GiteaMirror commented 2026-04-22 02:15:31 -05:00
Author
Owner
Copy Link

@tuoxin126 commented on GitHub (Oct 31, 2023):

Hi @tuoxin126 . Ollama should use the GPU by default. Here's a few things you can try:

  • Make sure nvidia drivers are installed. Running nvidia-smi should work.
  • Try running ollama run <model>. Does this use the GPU?

The Linux installer script will install Nvidia drivers if they aren't installed yet:

curl https://ollama.ai/install.sh | sh

If not it may be something we can look into further to help debug 😊 . I'll close this issue for now but please re-open if above doesn't help

GPU can be used through curl. Thank you for your reply

<!-- gh-comment-id:1786455211 --> @tuoxin126 commented on GitHub (Oct 31, 2023): > Hi @tuoxin126 . Ollama should use the GPU by default. Here's a few things you can try: > > * Make sure nvidia drivers are installed. Running `nvidia-smi` should work. > * Try running `ollama run <model>`. Does this use the GPU? > > The Linux installer script will install Nvidia drivers if they aren't installed yet: > > ``` > curl https://ollama.ai/install.sh | sh > ``` > > If not it may be something we can look into further to help debug 😊 . I'll close this issue for now but please re-open if above doesn't help GPU can be used through curl. Thank you for your reply
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 bug
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#26199
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?