github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11747] GPT-OSS:20b offloaded partially to CPU despite enough VRAM #33543

New Issue
Closed
opened 2026-04-22 16:23:01 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-22 16:23:01 -05:00
Owner
Copy Link

Originally created by @MarkEScheidker on GitHub (Aug 6, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11747

What is the issue?

I have a 16gb Nvidia GPU, but Ollama doesn't use all available VRAM unless I specify number of layers to use in OpenWebui.

Default settings:
Image

OpenWebUI num_gpu (ollama) set to 256:

Image

I've set these environment variables:
OLLAMA_FLASH_ATTENTION=1
OLLAMA_NUM_PARALLEL=1
GGML_USE_MMAP=1

Is there a reason the model isn't fully loaded into vram by default?

Relevant log output

OS

WSL2, Docker

GPU

Nvidia

CPU

Intel

Ollama version

0.11.2

Originally created by @MarkEScheidker on GitHub (Aug 6, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11747 ### What is the issue? I have a 16gb Nvidia GPU, but Ollama doesn't use all available VRAM unless I specify number of layers to use in OpenWebui. Default settings: <img width="501" height="780" alt="Image" src="https://github.com/user-attachments/assets/9b371b2b-308b-43f1-8b45-d08ad799724b" /> OpenWebUI num_gpu (ollama) set to 256: <img width="494" height="786" alt="Image" src="https://github.com/user-attachments/assets/bdc39144-2806-40b3-93c1-bc38482e3d6f" /> I've set these environment variables: OLLAMA_FLASH_ATTENTION=1 OLLAMA_NUM_PARALLEL=1 GGML_USE_MMAP=1 Is there a reason the model isn't fully loaded into vram by default? ### Relevant log output ```shell ``` ### OS WSL2, Docker ### GPU Nvidia ### CPU Intel ### Ollama version 0.11.2
GiteaMirror added the bug label 2026-04-22 16:23:01 -05:00
GiteaMirror commented 2026-04-22 16:23:03 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Aug 6, 2025):

Hi there. This should be fixed in 0.11.3 – let me know if you're still seeing issues!

<!-- gh-comment-id:3160807155 --> @jmorganca commented on GitHub (Aug 6, 2025): Hi there. This should be fixed in 0.11.3 – let me know if you're still seeing issues!
GiteaMirror commented 2026-04-22 16:23:05 -05:00
Author
Owner
Copy Link

@MarkEScheidker commented on GitHub (Aug 6, 2025):

Yes, It did, the latest tag for the docker image was just updated so I wasn't using the latest verison

<!-- gh-comment-id:3160967499 --> @MarkEScheidker commented on GitHub (Aug 6, 2025): Yes, It did, the latest tag for the docker image was just updated so I wasn't using the latest verison
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#33543
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?