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 #2385] ollama breaks running qwen on ubuntu 20 #27146

New Issue
Closed
opened 2026-04-22 04:07:26 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-22 04:07:26 -05:00
Owner
Copy Link

Originally created by @cognitivetech on GitHub (Feb 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2385

Either using the version included with ollama pull qwen or using my own custom modelfile with q8 and chatml template qwen causes ollama to get "stuck" it doesn't use GPU for qwen, or any other working model after trying qwen until reboot.

see also: https://github.com/ollama/ollama/issues/1691

Originally created by @cognitivetech on GitHub (Feb 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2385 Either using the version included with `ollama pull qwen` or using my own custom modelfile with q8 and chatml template qwen causes ollama to get "stuck" it doesn't use GPU for qwen, or any other working model after trying qwen until reboot. see also: https://github.com/ollama/ollama/issues/1691
GiteaMirror commented 2026-04-22 04:07:29 -05:00
Author
Owner
Copy Link

@cognitivetech commented on GitHub (Feb 7, 2024):

also this qwen template seems not right (https://github.com/ollama/ollama/issues/1977)

# Modelfile generated by "ollama show"
# To build a new Modelfile based on this one, replace the FROM line with:
# FROM qwen:latest

FROM /usr/share/ollama/.ollama/models/blobs/sha256:46bb65206e0e2b00424f33985a5281bd21070617ebcfda9be86eb17e6e00f793
TEMPLATE """{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>{{ end }}<|im_start|>user
{{ .Prompt }}<|im_end|>
<|im_start|>assistant
"""
PARAMETER stop "<|im_start|>"
PARAMETER stop "<|im_end|>"

for one thing, it's got no newlines, does not chatml have newline?
for another, I would guess that the "lm_start" parameter should read like this: PARAMETER start "<|im_start|>"

I don't have tons of time to test this right now, especially as it requires a reboot for each test. (I read somewhere a command to restart ollama service, but I can't find now). However, if someone wants to share a debug command so I can see what is actually happening, I can do that much.

<!-- gh-comment-id:1931765816 --> @cognitivetech commented on GitHub (Feb 7, 2024): also this qwen template seems not right (https://github.com/ollama/ollama/issues/1977) ``` # Modelfile generated by "ollama show" # To build a new Modelfile based on this one, replace the FROM line with: # FROM qwen:latest FROM /usr/share/ollama/.ollama/models/blobs/sha256:46bb65206e0e2b00424f33985a5281bd21070617ebcfda9be86eb17e6e00f793 TEMPLATE """{{ if .System }}<|im_start|>system {{ .System }}<|im_end|>{{ end }}<|im_start|>user {{ .Prompt }}<|im_end|> <|im_start|>assistant """ PARAMETER stop "<|im_start|>" PARAMETER stop "<|im_end|>" ``` for one thing, it's got no newlines, does not chatml have newline? for another, I would guess that the "lm_start" parameter should read like this: `PARAMETER start "<|im_start|>"` I don't have tons of time to test this right now, especially as it requires a reboot for each test. (I read somewhere a command to restart ollama service, but I can't find now). However, if someone wants to share a debug command so I can see what is actually happening, I can do that much.
GiteaMirror commented 2026-04-22 04:07:30 -05:00
Author
Owner
Copy Link

@cognitivetech commented on GitHub (Feb 9, 2024):

needed ollama update. but really if the model is not supported then it should not just hang forever.. but fail gracefully

<!-- gh-comment-id:1936580675 --> @cognitivetech commented on GitHub (Feb 9, 2024): needed ollama update. but really if the model is not supported then it should not just hang forever.. but fail gracefully
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
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#27146
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?