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 #10469] Running qwen3 returns an error #68943

New Issue
Closed
opened 2026-05-04 16:13:46 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-05-04 16:13:46 -05:00
Owner
Copy Link

Originally created by @Hunter6324 on GitHub (Apr 29, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10469

What is the issue?

PS C:\WINDOWS\system32> ollama run qwen3:14b
Error: llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed

Relevant log output

PS C:\WINDOWS\system32> ollama run qwen3:14b
Error: llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.6.6

Originally created by @Hunter6324 on GitHub (Apr 29, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10469 ### What is the issue? PS C:\WINDOWS\system32> ollama run qwen3:14b Error: llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed ### Relevant log output ```shell PS C:\WINDOWS\system32> ollama run qwen3:14b Error: llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.6.6
GiteaMirror added the bug label 2026-05-04 16:13:46 -05:00
GiteaMirror commented 2026-05-04 16:14:06 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 29, 2025):

#9509

<!-- gh-comment-id:2838250609 --> @rick-github commented on GitHub (Apr 29, 2025): #9509
GiteaMirror commented 2026-05-04 16:14:13 -05:00
Author
Owner
Copy Link

@Hunter6324 commented on GitHub (Apr 29, 2025):

#9509

Miracle doctor!

<!-- gh-comment-id:2838408095 --> @Hunter6324 commented on GitHub (Apr 29, 2025): > [#9509](https://github.com/ollama/ollama/issues/9509) Miracle doctor!
GiteaMirror commented 2026-05-04 16:14:23 -05:00
Author
Owner
Copy Link

@Hunter6324 commented on GitHub (Apr 29, 2025):

GitHub Copilot:

This issue might be caused by a conflict between Ollama and certain library files (DLL files) installed by Docker Desktop.

The idea behind the solution is to find these potentially conflicting Docker files and rename them so that Ollama doesn't load them.

Here are the specific steps:

Open Command Prompt or PowerShell as Administrator:

Type cmd or powershell in the Windows search bar.
Right-click on "Command Prompt" or "Windows PowerShell" in the search results.
Select "Run as administrator".
Change to the Docker Directory:

In the administrator Command Prompt or PowerShell window that opens, type the following command and press Enter:
cd "C:\Program Files\Docker\Docker\resources\bin"
Rename the Conflicting DLL Files:

Execute the following commands one by one to rename each file (adding a .old suffix to the original filename):
ren ggml-base.dll ggml-base.dll.old
ren ggml-cpu.dll ggml-cpu.dll.old
ren ggml.dll ggml.dll.old
ren llama.dll llama.dll.old
If a file doesn't exist, you'll see an error message. You can ignore it and proceed to the next command.
Retry Running the Ollama Command:

Go back to the terminal window where you were previously running Ollama (or run it directly in the administrator window) and try again:
ollama run qwen3:14b
Note:

Renaming these files might affect some functionalities of Docker Desktop. If you encounter issues with Docker Desktop, you can rename the files back (remove the .old suffix).
This method is based on the experience of other users and is not guaranteed to solve your problem, but it's worth trying.

<!-- gh-comment-id:2838430227 --> @Hunter6324 commented on GitHub (Apr 29, 2025): GitHub Copilot: This issue might be caused by a conflict between Ollama and certain library files (DLL files) installed by Docker Desktop. The idea behind the solution is to find these potentially conflicting Docker files and rename them so that Ollama doesn't load them. Here are the specific steps: Open Command Prompt or PowerShell as Administrator: Type cmd or powershell in the Windows search bar. Right-click on "Command Prompt" or "Windows PowerShell" in the search results. Select "Run as administrator". Change to the Docker Directory: In the administrator Command Prompt or PowerShell window that opens, type the following command and press Enter: cd "C:\Program Files\Docker\Docker\resources\bin" Rename the Conflicting DLL Files: Execute the following commands one by one to rename each file (adding a .old suffix to the original filename): ren ggml-base.dll ggml-base.dll.old ren ggml-cpu.dll ggml-cpu.dll.old ren ggml.dll ggml.dll.old ren llama.dll llama.dll.old If a file doesn't exist, you'll see an error message. You can ignore it and proceed to the next command. Retry Running the Ollama Command: Go back to the terminal window where you were previously running Ollama (or run it directly in the administrator window) and try again: ollama run qwen3:14b Note: Renaming these files might affect some functionalities of Docker Desktop. If you encounter issues with Docker Desktop, you can rename the files back (remove the .old suffix). This method is based on the experience of other users and is not guaranteed to solve your problem, but it's worth trying.
GiteaMirror commented 2026-05-04 16:14:29 -05:00
Author
Owner
Copy Link

@bjorn-ver commented on GitHub (Apr 29, 2025):

This seems to be te problem. Docker desktop is updates and caused te problem.
I've checked the path, Ollama is first in the path string. So why would it take a dll from some unrelated folder?
This is not a solution but a workarround, it still needs a bug fix.

<!-- gh-comment-id:2839328970 --> @bjorn-ver commented on GitHub (Apr 29, 2025): This seems to be te problem. Docker desktop is updates and caused te problem. I've checked the path, Ollama is first in the path string. So why would it take a dll from some unrelated folder? This is not a solution but a workarround, it still needs a bug fix.
GiteaMirror commented 2026-05-04 16:14:33 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 29, 2025):

Set OLLAMA_DEBUG=1 and the logs will show that ollama does exclude paths not directly related to ollama. This might be some OS level activity. #9509 is tracking this issue.

<!-- gh-comment-id:2839387736 --> @rick-github commented on GitHub (Apr 29, 2025): Set `OLLAMA_DEBUG=1` and the logs will show that ollama does exclude paths not directly related to ollama. This might be some OS level activity. #9509 is tracking this issue.
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#68943
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?