github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7996] Less available memory than expected #67172

New Issue
Closed
opened 2026-05-04 09:34:31 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-05-04 09:34:31 -05:00
Owner
Copy Link

Originally created by @tie-pilot-qxw on GitHub (Dec 8, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7996

What is the issue?

I'm using a 4060 laptop GPU, which has 8GB of memory. However, from the log, it showed that
library=cuda variant=v12 compute=8.9 driver=12.7 name="NVIDIA GeForce RTX 4060 Laptop GPU" total="8.0 GiB" available="6.9 GiB"
That's quite strange for from nvidia-smi
image
Actually, when I tried to run llama3.2 3b fp16, ollama offloads 10% of the layers to the CPU, while from ollama ps, the needed memory is 7.9GB, which should be able to fit in the GPU memory

OS

WSL2

GPU

Nvidia

CPU

Intel

Ollama version

0.4.7

Originally created by @tie-pilot-qxw on GitHub (Dec 8, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7996 ### What is the issue? I'm using a 4060 laptop GPU, which has 8GB of memory. However, from the log, it showed that `library=cuda variant=v12 compute=8.9 driver=12.7 name="NVIDIA GeForce RTX 4060 Laptop GPU" total="8.0 GiB" available="6.9 GiB"` That's quite strange for from `nvidia-smi` ![image](https://github.com/user-attachments/assets/f7e61eb4-ee66-4927-aa71-295a095c5bb0) Actually, when I tried to run llama3.2 3b fp16, ollama offloads 10% of the layers to the CPU, while from `ollama ps`, the needed memory is 7.9GB, which should be able to fit in the GPU memory ### OS WSL2 ### GPU Nvidia ### CPU Intel ### Ollama version 0.4.7
GiteaMirror added the bugneeds more info labels 2026-05-04 09:34:31 -05:00
GiteaMirror commented 2026-05-04 09:34:33 -05:00
Author
Owner
Copy Link

@fxmbsw7 commented on GitHub (Dec 10, 2024):

disable hard apps in autostart , disable graphica bling blinng , reboot - to clear memory and start less gpu mem eating apps
then try ollama

<!-- gh-comment-id:2530585520 --> @fxmbsw7 commented on GitHub (Dec 10, 2024): disable hard apps in autostart , disable graphica bling blinng , reboot - to clear memory and start less gpu mem eating apps then try ollama
GiteaMirror commented 2026-05-04 09:34:33 -05:00
Author
Owner
Copy Link

@fxmbsw7 commented on GitHub (Dec 10, 2024):

try if u can drive ur laptop or monitor via igpu , and leave the nv gpu only to ollama

<!-- gh-comment-id:2530586581 --> @fxmbsw7 commented on GitHub (Dec 10, 2024): try if u can drive ur laptop or monitor via igpu , and leave the nv gpu only to ollama
GiteaMirror commented 2026-05-04 09:34:33 -05:00
Author
Owner
Copy Link

@tie-pilot-qxw commented on GitHub (Dec 10, 2024):

try if u can drive ur laptop or monitor via igpu , and leave the nv gpu only to ollama

I have set the laptop only to use the igup, however, ollama still shows that only 6.9GB is available while both the task manager and the nvidia-smi tell me all the 8GB are available

<!-- gh-comment-id:2530668835 --> @tie-pilot-qxw commented on GitHub (Dec 10, 2024): > try if u can drive ur laptop or monitor via igpu , and leave the nv gpu only to ollama I have set the laptop only to use the igup, however, ollama still shows that only 6.9GB is available while both the task manager and the nvidia-smi tell me all the 8GB are available
GiteaMirror commented 2026-05-04 09:34:34 -05:00
Author
Owner
Copy Link

@fxmbsw7 commented on GitHub (Dec 10, 2024):

stuff seems using it

1 look in nv tool ( i dunno name i dont have nvidia since long )
what it says

2 go windows settings > system > display
in one of the two middle tabs
theres apps to select

u got windows btw ?
select all power saver
.. maybe add ollama that ollama that runs
process viewer > details , rightclick on the ollama and in menu should be '
go to directory '
then add this ollama as performance

50 50 chance

On Tue, Dec 10, 2024, 8:33 AM qiangxinwei @.***> wrote:

try if u can drive ur laptop or monitor via igpu , and leave the nv gpu
only to ollama

I have set the laptop only to use the igup, however, ollama still shows
that only 6.9GB is available while both the task manager and the nvidia-smi
tell me all the 8GB are available

—
Reply to this email directly, view it on GitHub
https://github.com/ollama/ollama/issues/7996#issuecomment-2530668835,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AJMLP3MG73PE5GCVGRMDGRT2E2KMTAVCNFSM6AAAAABTG4MBCGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMZQGY3DQOBTGU
.
You are receiving this because you commented.Message ID:
@.***>

<!-- gh-comment-id:2530678852 --> @fxmbsw7 commented on GitHub (Dec 10, 2024): stuff seems using it 1 look in nv tool ( i dunno name i dont have nvidia since long ) what it says 2 go windows settings > system > display in one of the two middle tabs theres apps to select u got windows btw ? select all power saver .. maybe add ollama that ollama that runs process viewer > details , rightclick on the ollama and in menu should be ' go to directory ' then add this ollama as performance 50 50 chance On Tue, Dec 10, 2024, 8:33 AM qiangxinwei ***@***.***> wrote: > try if u can drive ur laptop or monitor via igpu , and leave the nv gpu > only to ollama > > I have set the laptop only to use the igup, however, ollama still shows > that only 6.9GB is available while both the task manager and the nvidia-smi > tell me all the 8GB are available > > — > Reply to this email directly, view it on GitHub > <https://github.com/ollama/ollama/issues/7996#issuecomment-2530668835>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AJMLP3MG73PE5GCVGRMDGRT2E2KMTAVCNFSM6AAAAABTG4MBCGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMZQGY3DQOBTGU> > . > You are receiving this because you commented.Message ID: > ***@***.***> >
GiteaMirror commented 2026-05-04 09:34:34 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Dec 14, 2024):

Server logs and the output of nvidia-smi -q may aid in debugging.

<!-- gh-comment-id:2543175961 --> @rick-github commented on GitHub (Dec 14, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) and the output of `nvidia-smi -q` may aid in debugging.
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 needs more info
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#67172
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?