github-starred/ollama-ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-27 07:57:18 -05:00
Code Issues 2.2k Packages Projects Releases 100 Wiki Activity

Intel Iris Xe Graphics (16GB) not detected by Ollama v0.12.10 on Windows 11 despite Vulkan/DXGI+PDH support #8633

New Issue
Open
opened 2025-11-12 14:47:54 -06:00 by GiteaMirror · 3 comments
GiteaMirror commented 2025-11-12 14:47:54 -06:00
Owner
Copy Link

Originally created by @deep1305 on GitHub (Nov 9, 2025).

What is the issue?

Ollama v0.12.10 fails to detect my Intel Iris Xe Graphics (integrated GPU with 16GB shared memory) on Windows 11, despite the changelog mentioning "Add Vulkan memory detection for Intel GPU using DXGI+PDH". The system falls back to 100% CPU mode with total vram="0 B" and offloaded 0/49 layers to GPU.

Vulkan is properly installed and vulkaninfo correctly lists the Intel GPU, but Ollama never loads a Vulkan backend library and only uses CPU.

Expected Behavior

  • Intel Iris Xe should be detected via Vulkan/DXGI+PDH
  • Model layers should offload to GPU
  • ollama ps should show GPU percentage

Actual Behavior

  • 100% CPU mode
  • total vram="0 B"
  • offloaded 0/49 layers to GPU
  • No Vulkan backend library loaded

Relevant log output

time=2025-11-06T18:33:02.776-05:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2025-11-06T18:33:02.776-05:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="31.7 GiB" available="23.1 GiB"
time=2025-11-06T18:33:02.776-05:00 level=INFO source=routes.go:1619 msg="entering low vram mode" "total vram"="0 B" threshold="20.0 GiB"


load_backend: loaded CPU backend from C:\Users\smart\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll

time=2025-11-06T18:33:10.143-05:00 level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU"
time=2025-11-06T18:33:10.144-05:00 level=INFO source=ggml.go:486 msg="offloading output layer to CPU"
time=2025-11-06T18:33:10.144-05:00 level=INFO source=ggml.go:494 msg="offloaded 0/49 layers to GPU"

OS

Windows

GPU

Intel

CPU

No response

Ollama version

0.12.10

Originally created by @deep1305 on GitHub (Nov 9, 2025). ### What is the issue? Ollama v0.12.10 fails to detect my Intel Iris Xe Graphics (integrated GPU with 16GB shared memory) on Windows 11, despite the changelog mentioning "Add Vulkan memory detection for Intel GPU using DXGI+PDH". The system falls back to 100% CPU mode with total vram="0 B" and offloaded 0/49 layers to GPU. Vulkan is properly installed and vulkaninfo correctly lists the Intel GPU, but Ollama never loads a Vulkan backend library and only uses CPU. ## Expected Behavior - Intel Iris Xe should be detected via Vulkan/DXGI+PDH - Model layers should offload to GPU - `ollama ps` should show GPU percentage ## Actual Behavior - 100% CPU mode - total vram="0 B" - offloaded 0/49 layers to GPU - No Vulkan backend library loaded ### Relevant log output ```shell time=2025-11-06T18:33:02.776-05:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2025-11-06T18:33:02.776-05:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="31.7 GiB" available="23.1 GiB" time=2025-11-06T18:33:02.776-05:00 level=INFO source=routes.go:1619 msg="entering low vram mode" "total vram"="0 B" threshold="20.0 GiB" load_backend: loaded CPU backend from C:\Users\smart\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll time=2025-11-06T18:33:10.143-05:00 level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU" time=2025-11-06T18:33:10.144-05:00 level=INFO source=ggml.go:486 msg="offloading output layer to CPU" time=2025-11-06T18:33:10.144-05:00 level=INFO source=ggml.go:494 msg="offloaded 0/49 layers to GPU" ``` ### OS Windows ### GPU Intel ### CPU _No response_ ### Ollama version 0.12.10
GiteaMirror added the bugneeds more info labels 2025-11-12 14:47:54 -06:00
GiteaMirror commented 2025-11-12 14:47:55 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 9, 2025):

The Vulkan backend is not yet enabled in production releases. If you want to use the Vulkan backend, install the Vulkan SDK and set VULKAN_SDK in your environment, then follow the developer instructions. In a future release, Vulkan support will be included in the binary release as well. Please file issues if you run into any problems.

@rick-github commented on GitHub (Nov 9, 2025): The Vulkan backend is not yet enabled in production releases. If you want to use the Vulkan backend, install the [Vulkan SDK](https://vulkan.lunarg.com/) and set VULKAN_SDK in your environment, then follow the [developer instructions](https://github.com/ollama/ollama/blob/main/docs/development.md). In a future release, Vulkan support will be included in the binary release as well. Please file issues if you run into any problems.
GiteaMirror commented 2025-11-12 14:47:55 -06:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Nov 12, 2025):

@deep1305 were you able to get it to work? We will enable it by default soon; still trying to get people to try it out and report bugs.

@pdevine commented on GitHub (Nov 12, 2025): @deep1305 were you able to get it to work? We will enable it by default soon; still trying to get people to try it out and report bugs.
GiteaMirror commented 2025-11-12 14:47:56 -06:00
Author
Owner
Copy Link

@ndragon798 commented on GitHub (Nov 12, 2025):

Hey @pdevine I just tested building and running using the latest vulkan sdk and my intel B50. Everything works great on Fedora 42

time=2025-11-12T10:10:42.881-05:00 level=INFO source=routes.go:1597 msg="Listening on 127.0.0.1:11434 (version 0.0.0)"
time=2025-11-12T10:10:42.882-05:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2025-11-12T10:10:42.882-05:00 level=INFO source=server.go:392 msg="starting runner" cmd="/home/nathan/.cache/go-build/16/16d3b3dc5dca6bf7f584731b8b542153ff61e0c555ff1cd863327f882730942c-d/ollama runner --ollama-engine --port 41727"
time=2025-11-12T10:10:44.817-05:00 level=INFO source=types.go:42 msg="inference compute" id=868012e2-0000-0000-0e00-000000000000 filter_id="" library=Vulkan compute=0.0 name=Vulkan0 description="Intel(R) Arc(tm) Pro B50 Graphics (BMG G21)" libdirs=ollama driver=0.0 pci_id=0000:0e:00.0 type=discrete total="15.9 GiB" available="12.9 GiB"
time=2025-11-12T10:10:44.818-05:00 level=INFO source=routes.go:1638 msg="entering low vram mode" "total vram"="15.9 GiB" threshold="20.0 GiB"

Is there anything you want tested?

@ndragon798 commented on GitHub (Nov 12, 2025): Hey @pdevine I just tested building and running using the latest vulkan sdk and my intel B50. Everything works great on Fedora 42 ``` time=2025-11-12T10:10:42.881-05:00 level=INFO source=routes.go:1597 msg="Listening on 127.0.0.1:11434 (version 0.0.0)" time=2025-11-12T10:10:42.882-05:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2025-11-12T10:10:42.882-05:00 level=INFO source=server.go:392 msg="starting runner" cmd="/home/nathan/.cache/go-build/16/16d3b3dc5dca6bf7f584731b8b542153ff61e0c555ff1cd863327f882730942c-d/ollama runner --ollama-engine --port 41727" time=2025-11-12T10:10:44.817-05:00 level=INFO source=types.go:42 msg="inference compute" id=868012e2-0000-0000-0e00-000000000000 filter_id="" library=Vulkan compute=0.0 name=Vulkan0 description="Intel(R) Arc(tm) Pro B50 Graphics (BMG G21)" libdirs=ollama driver=0.0 pci_id=0000:0e:00.0 type=discrete total="15.9 GiB" available="12.9 GiB" time=2025-11-12T10:10:44.818-05:00 level=INFO source=routes.go:1638 msg="entering low vram mode" "total vram"="15.9 GiB" threshold="20.0 GiB" ``` Is there anything you want tested?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
client2
cloud
compatibility
context-length
create
docker
documentation
embeddings
engine
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
install
integration
intel
js
linux
macos
memory
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
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-ollama#8633
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?