github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-21 05:01:48 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #6238] Ollama server running out of memory when it didn't in previous version #81582

New Issue
Closed
opened 2026-05-09 11:30:13 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-05-09 11:30:13 -05:00
Owner
Copy Link

Originally created by @MxtAppz on GitHub (Aug 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6238

What is the issue?

Hello,

I'm trying to run Llama 3.1 8b in Ollama 0.3.4 on a laptop (it has 8gb ram and 4 cpu cores, running it on CPU as my GPU is integrated and not compatible). Maybe this sounds crazy, but it worked fine on Ollama 0.3.3, and just after updating to 0.3.4 and running it (I also tried updating the model but the same) I get:

Error: Post "http://127.0.0.1:11434/api/chat": EOF

and when I try again:

Error: llama runner process has terminated: signal: killed

So I think it's running out of memory. I like to try things on my laptop before deploying them to my server, so I would like if you could "fix" that back, or show me a workaround for this. (Mistral 7b works fine with 0.3.4).

Thanks in advanced.

OS

Linux

GPU

Other

CPU

Intel

Ollama version

0.3.4

Originally created by @MxtAppz on GitHub (Aug 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6238 ### What is the issue? Hello, I'm trying to run Llama 3.1 8b in Ollama 0.3.4 on a laptop (it has 8gb ram and 4 cpu cores, running it on CPU as my GPU is integrated and not compatible). Maybe this sounds crazy, but it worked fine on Ollama 0.3.3, and just after updating to 0.3.4 and running it (I also tried updating the model but the same) I get: Error: Post "http://127.0.0.1:11434/api/chat": EOF and when I try again: Error: llama runner process has terminated: signal: killed So I think it's running out of memory. I like to try things on my laptop before deploying them to my server, so I would like if you could "fix" that back, or show me a workaround for this. (Mistral 7b works fine with 0.3.4). Thanks in advanced. ### OS Linux ### GPU Other ### CPU Intel ### Ollama version 0.3.4
GiteaMirror added the bug label 2026-05-09 11:30:13 -05:00
GiteaMirror commented 2026-05-09 11:30:17 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Aug 7, 2024):

Server logs will help in debugging.

<!-- gh-comment-id:2273966029 --> @rick-github commented on GitHub (Aug 7, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will help in debugging.
GiteaMirror commented 2026-05-09 11:30:18 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Aug 7, 2024):

@MxtAppz Alongside the new version of Ollama, there is an update for the Llama3.1 LLMs. Did you pull this latest version? If you are experiencing memory issues, could you try a lighter version of Llama3.1?

ollama run llama3.1:8b-instruct-q3_K_L

This version is slightly lighter than the default q4_0 and should resolve the issue you're facing.

Yes, the system size grows as Ollama and Llama3.1 grow as well. Another option is to restart your Mac before testing Ollama and ensure no other applications are running besides Ollama and the terminal.

<!-- gh-comment-id:2273988758 --> @igorschlum commented on GitHub (Aug 7, 2024): @MxtAppz Alongside the new version of Ollama, there is an update for the Llama3.1 LLMs. Did you pull this latest version? If you are experiencing memory issues, could you try a lighter version of Llama3.1? ```bash ollama run llama3.1:8b-instruct-q3_K_L ``` This version is slightly lighter than the default q4_0 and should resolve the issue you're facing. Yes, the system size grows as Ollama and Llama3.1 grow as well. Another option is to restart your Mac before testing Ollama and ensure no other applications are running besides Ollama and the terminal.
GiteaMirror commented 2026-05-09 11:30:18 -05:00
Author
Owner
Copy Link

@MxtAppz commented on GitHub (Aug 7, 2024):

@MxtAppz Alongside the new version of Ollama, there is an update for the Llama3.1 LLMs. Did you pull this latest version? If you are experiencing memory issues, could you try a lighter version of Llama3.1?

ollama run llama3.1:8b-instruct-q3_K_L

This version is slightly lighter than the default q4_0 and should resolve the issue you're facing.

Yes, the system size grows as Ollama and Llama3.1 grow as well. Another option is to restart your Mac before testing Ollama and ensure no other applications are running besides Ollama and the terminal.

Yes, I pulled the latest llama3.1 model, and doesn't work (the older one didn't work either though). I'll try the instruct version as soon as possible and tell you.

<!-- gh-comment-id:2274005462 --> @MxtAppz commented on GitHub (Aug 7, 2024): > @MxtAppz Alongside the new version of Ollama, there is an update for the Llama3.1 LLMs. Did you pull this latest version? If you are experiencing memory issues, could you try a lighter version of Llama3.1? > > ```shell > ollama run llama3.1:8b-instruct-q3_K_L > ``` > > This version is slightly lighter than the default q4_0 and should resolve the issue you're facing. > > Yes, the system size grows as Ollama and Llama3.1 grow as well. Another option is to restart your Mac before testing Ollama and ensure no other applications are running besides Ollama and the terminal. Yes, I pulled the latest llama3.1 model, and doesn't work (the older one didn't work either though). I'll try the instruct version as soon as possible and tell you.
GiteaMirror commented 2026-05-09 11:30:19 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Aug 7, 2024):

Error: llama runner process has terminated: signal: killed

This indicates a problem with the runner, not the model (unless the model is borked somehow). Server logs will help in debugging.

<!-- gh-comment-id:2274030148 --> @rick-github commented on GitHub (Aug 7, 2024): ``` Error: llama runner process has terminated: signal: killed ``` This indicates a problem with the runner, not the model (unless the model is borked somehow). Server logs will help in debugging.
GiteaMirror commented 2026-05-09 11:30:19 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Aug 7, 2024):

@MxtAppz would it be possible to check the logs with journalctl -u ollama to see what the error may be? Sorry about this

<!-- gh-comment-id:2274039538 --> @jmorganca commented on GitHub (Aug 7, 2024): @MxtAppz would it be possible to check the logs with `journalctl -u ollama` to see what the error may be? Sorry about this
GiteaMirror commented 2026-05-09 11:30:20 -05:00
Author
Owner
Copy Link

@MxtAppz commented on GitHub (Aug 7, 2024):

Hello newly,

Finally, the issue was resolved by rebooting the system (thanks to @igorschlum for telling me). Anyway, I will leave the issue open and attach the log after updating Ollama in case someone wants to check it out. Thanks for your time.
ollama.txt

<!-- gh-comment-id:2274371118 --> @MxtAppz commented on GitHub (Aug 7, 2024): Hello newly, Finally, the issue was resolved by rebooting the system (thanks to @igorschlum for telling me). Anyway, I will leave the issue open and attach the log after updating Ollama in case someone wants to check it out. Thanks for your time. [ollama.txt](https://github.com/user-attachments/files/16534385/ollama.txt)
GiteaMirror commented 2026-05-09 11:30:20 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Aug 13, 2024):

This should be fixed now – you'll need to re-pull llama3.1 let me know if that doesn't help!

<!-- gh-comment-id:2285383932 --> @jmorganca commented on GitHub (Aug 13, 2024): This should be fixed now – you'll need to re-pull `llama3.1` let me know if that doesn't help!
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#81582
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?