github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4223] qwen:72b-chat-q4_K_S does not load #49144

New Issue
Closed
opened 2026-04-28 10:49:36 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-28 10:49:36 -05:00
Owner
Copy Link

Originally created by @saddy001 on GitHub (May 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4223

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

Where the model "qwen:72b" loads successfully, the model "qwen:72b-chat-q4_K_S" does not load. The loading spinner just doesn't stop even after waiting a long time. Since the models occupy the same amount of memory (41 GB) I assume the RAM usage is roughly the same. Can somebody reproduce this?

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.33

Originally created by @saddy001 on GitHub (May 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4223 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? Where the model "qwen:72b" loads successfully, the model "qwen:72b-chat-q4_K_S" does not load. The loading spinner just doesn't stop even after waiting a long time. Since the models occupy the same amount of memory (41 GB) I assume the RAM usage is roughly the same. Can somebody reproduce this? ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.33
GiteaMirror added the memorybug labels 2026-04-28 10:49:36 -05:00
GiteaMirror commented 2026-04-28 10:49:38 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Jul 25, 2024):

It looks like our memory prediction is incorrect on this model. On a 48G GPU, we think we can load all 81 layers but it crashes with OOM. In reality, only ~60 layers will fit. Until we can get this fixed, you can specify an explicit num_gpu setting.

<!-- gh-comment-id:2251158439 --> @dhiltgen commented on GitHub (Jul 25, 2024): It looks like our memory prediction is incorrect on this model. On a 48G GPU, we think we can load all 81 layers but it crashes with OOM. In reality, only ~60 layers will fit. Until we can get this fixed, you can specify an explicit `num_gpu` setting.
GiteaMirror commented 2026-04-28 10:49:40 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Sep 24, 2025):

I'm going to go ahead and close this now that the new memory management logic is on by default. If you continue to see problems, please file a new issue.

<!-- gh-comment-id:3330124977 --> @jessegross commented on GitHub (Sep 24, 2025): I'm going to go ahead and close this now that the new memory management logic is on by default. If you continue to see problems, please file a new 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 memory
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#49144
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?