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 #9167] ollama start log output "detected OS VRAM overhead" #68023

New Issue
Closed
opened 2026-05-04 12:16:34 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 12:16:34 -05:00
Owner
Copy Link

Originally created by @chinafuxi on GitHub (Feb 17, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9167

What is the issue?

my compute hava 128G RAM and 22GB VRAM, but when i run 32b model,the log output this log
time=2025-02-17T16:52:12.068+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-66fb2fe2-3f29-15f8-22fe-2670176a5d13 library=cuda compute=7.5 driver=12.8 name="NVIDIA GeForce RTX 2080 Ti" overhead="3.2 GiB"

then run 32b model , ollama can be used 18.3GB VRAM, should be use 20.2GB VRAM,system occupies 3.2GB,some models are load to memory,RAM uesd 2.1GB

time=2025-02-17T16:52:12.221+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=65 layers.offload=59 layers.split="" memory.available="[18.3 GiB]" memory.gpu_overhead="0 B" memory.required.full="20.2 GiB"

Image

my os version info
OS Name: Microsoft Windows 10 Pro
OS Version: 10.0.18363 N/A Build 18363

this is stop allama VRAM usage

Image

Relevant log output

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.5.9

Originally created by @chinafuxi on GitHub (Feb 17, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9167 ### What is the issue? my compute hava 128G RAM and 22GB VRAM, but when i run 32b model,the log output this log time=2025-02-17T16:52:12.068+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-66fb2fe2-3f29-15f8-22fe-2670176a5d13 library=cuda compute=7.5 driver=12.8 name="NVIDIA GeForce RTX 2080 Ti" overhead="3.2 GiB" then run 32b model , ollama can be used 18.3GB VRAM, should be use 20.2GB VRAM,system occupies 3.2GB,some models are load to memory,RAM uesd 2.1GB time=2025-02-17T16:52:12.221+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=65 layers.offload=59 layers.split="" memory.available="[18.3 GiB]" memory.gpu_overhead="0 B" memory.required.full="20.2 GiB" ![Image](https://github.com/user-attachments/assets/32d842f1-b9e2-4311-8824-8996c9d2695c) my os version info OS Name: Microsoft Windows 10 Pro OS Version: 10.0.18363 N/A Build 18363 this is stop allama VRAM usage ![Image](https://github.com/user-attachments/assets/fcfa3f20-6558-4e02-b07c-f642e857f59f) ### Relevant log output ```shell ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.5.9
GiteaMirror added the bug label 2026-05-04 12:16:34 -05:00
GiteaMirror commented 2026-05-04 12:16:42 -05:00
Author
Owner
Copy Link

@chinafuxi commented on GitHub (Feb 19, 2025):

i find the problem in microsoft forum, this is win10 WDDM problem , os about 15%VRAM isreserved for os

https://answers.microsoft.com/en-us/windows/forum/all/windows-10-does-not-let-cuda-applications-to-use/cffb3fcd-5a21-46cf-8123-aa53bb8bafd6

<!-- gh-comment-id:2667259756 --> @chinafuxi commented on GitHub (Feb 19, 2025): i find the problem in microsoft forum, this is win10 WDDM problem , os about 15%VRAM isreserved for os https://answers.microsoft.com/en-us/windows/forum/all/windows-10-does-not-let-cuda-applications-to-use/cffb3fcd-5a21-46cf-8123-aa53bb8bafd6
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#68023
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?