github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 20:54:54 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #987] segmentation fault with prompts longer than 5 / 6 tokens on intel mac #46992

New Issue
Closed
opened 2026-04-28 02:31:12 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-28 02:31:12 -05:00
Owner
Copy Link

Originally created by @Serpico84 on GitHub (Nov 3, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/987

I'm running Ollama on a 2019 intel MacBook Pro with 32gb of RAM and a 4gb AMD GPU. macOS Monterey.
For some reson, every prompt longer than a few words on both codellama:7b and llama2:7b end up with Error: llama runner exited, you may not have enough available memory to run this model

Very short prompts work ok.

This is the server log file:

llama_new_context_with_model: n_ctx      = 2048
llama_new_context_with_model: freq_base  = 1000000.0
llama_new_context_with_model: freq_scale = 1
llama_new_context_with_model: kv self size  = 1024.00 MB
llama_new_context_with_model: compute buffer total size = 162.13 MB
llama server listening at http://127.0.0.1:49879

{"timestamp":1699023783,"level":"INFO","function":"main","line":1749,"message":"HTTP server listening","hostname":"127.0.0.1","port":49879}
{"timestamp":1699023783,"level":"INFO","function":"log_server_request","line":1240,"message":"request","remote_addr":"127.0.0.1","remote_port":50344,"status":200,"method":"HEAD","path":"/","params":{}}
2023/11/03 16:03:03 llama.go:442: llama runner started in 1.001631 seconds
[GIN] 2023/11/03 - 16:03:03 | 200 |  1.143871095s |       127.0.0.1 | POST     "/api/generate"
{"timestamp":1699023810,"level":"INFO","function":"log_server_request","line":1240,"message":"request","remote_addr":"127.0.0.1","remote_port":50346,"status":200,"method":"HEAD","path":"/","params":{}}
2023/11/03 16:03:30 llama.go:385: signal: segmentation fault
2023/11/03 16:03:30 llama.go:459: llama runner stopped successfully
[GIN] 2023/11/03 - 16:03:30 | 200 |  227.020112ms |       127.0.0.1 | POST     "/api/generate"

Not sure what can I do to tackle this problem or provide any more info, as far as I know I should be able to run those models just fine with my machine.

Originally created by @Serpico84 on GitHub (Nov 3, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/987 I'm running Ollama on a 2019 intel MacBook Pro with 32gb of RAM and a 4gb AMD GPU. macOS Monterey. For some reson, every prompt longer than a few words on both codellama:7b and llama2:7b end up with `Error: llama runner exited, you may not have enough available memory to run this model` Very short prompts work ok. This is the server log file: ``` llama_new_context_with_model: n_ctx = 2048 llama_new_context_with_model: freq_base = 1000000.0 llama_new_context_with_model: freq_scale = 1 llama_new_context_with_model: kv self size = 1024.00 MB llama_new_context_with_model: compute buffer total size = 162.13 MB llama server listening at http://127.0.0.1:49879 {"timestamp":1699023783,"level":"INFO","function":"main","line":1749,"message":"HTTP server listening","hostname":"127.0.0.1","port":49879} {"timestamp":1699023783,"level":"INFO","function":"log_server_request","line":1240,"message":"request","remote_addr":"127.0.0.1","remote_port":50344,"status":200,"method":"HEAD","path":"/","params":{}} 2023/11/03 16:03:03 llama.go:442: llama runner started in 1.001631 seconds [GIN] 2023/11/03 - 16:03:03 | 200 | 1.143871095s | 127.0.0.1 | POST "/api/generate" {"timestamp":1699023810,"level":"INFO","function":"log_server_request","line":1240,"message":"request","remote_addr":"127.0.0.1","remote_port":50346,"status":200,"method":"HEAD","path":"/","params":{}} 2023/11/03 16:03:30 llama.go:385: signal: segmentation fault 2023/11/03 16:03:30 llama.go:459: llama runner stopped successfully [GIN] 2023/11/03 - 16:03:30 | 200 | 227.020112ms | 127.0.0.1 | POST "/api/generate" ``` Not sure what can I do to tackle this problem or provide any more info, as far as I know I should be able to run those models just fine with my machine.
GiteaMirror commented 2026-04-28 02:31:14 -05:00
Author
Owner
Copy Link

@calebrcannon commented on GitHub (Nov 4, 2023):

I have the same issue

<!-- gh-comment-id:1793356158 --> @calebrcannon commented on GitHub (Nov 4, 2023): I have the same issue
GiteaMirror commented 2026-04-28 02:31:15 -05:00
Author
Owner
Copy Link

@gregahren commented on GitHub (Nov 4, 2023):

Same problem here.

<!-- gh-comment-id:1793434861 --> @gregahren commented on GitHub (Nov 4, 2023): Same problem here.
GiteaMirror commented 2026-04-28 02:31:15 -05:00
Author
Owner
Copy Link

@FetchFast commented on GitHub (Nov 6, 2023):

Same here, even with orca-mini. Just to confirm, "bin/ollama-runner: bad CPU type in executable" is irrelevant?

But llama2 with ID fe938a131f40 works for me.

I have a 2015 MBP, 8GB ram though. Maybe I've got no GPU to confuse it...?

<!-- gh-comment-id:1797050763 --> @FetchFast commented on GitHub (Nov 6, 2023): Same here, even with orca-mini. Just to confirm, "bin/ollama-runner: bad CPU type in executable" is irrelevant? But llama2 with ID fe938a131f40 works for me. I have a 2015 MBP, 8GB ram though. Maybe I've got no GPU to confuse it...?
GiteaMirror commented 2026-04-28 02:31:16 -05:00
Author
Owner
Copy Link

@sizhky commented on GitHub (Nov 7, 2023):

But llama2 with ID fe938a131f40 works for me.

I'm using the default llama2 model. Does it have this specific id? If not how do I run this specific ID?

<!-- gh-comment-id:1797731993 --> @sizhky commented on GitHub (Nov 7, 2023): > But llama2 with ID fe938a131f40 works for me. I'm using the default llama2 model. Does it have this specific id? If not how do I run this specific ID?
GiteaMirror commented 2026-04-28 02:31:16 -05:00
Author
Owner
Copy Link

@bgokden commented on GitHub (Nov 8, 2023):

Same issue. Is there a recommended solution?

<!-- gh-comment-id:1802821514 --> @bgokden commented on GitHub (Nov 8, 2023): Same issue. Is there a recommended solution?
GiteaMirror commented 2026-04-28 02:31:17 -05:00
Author
Owner
Copy Link

@orlyandico commented on GitHub (Nov 17, 2023):

It seems my bug report is a duplicate of this. I was getting the same error on my Mac Pro 2013 with Monterey 12.7.1 - but not on my MBP16 2019 with Ventura 13.4.1(c) which does have an Intel processor also.

So it appears to be OS-related.

I'll try to OCLP the trashcan to Ventura and see if this error disappears... would appreciate if the other folks with this error (and with natively Ventura-capable Macs) also test.

<!-- gh-comment-id:1815579415 --> @orlyandico commented on GitHub (Nov 17, 2023): It seems my bug report is a duplicate of this. I was getting the same error on my Mac Pro 2013 with Monterey 12.7.1 - but not on my MBP16 2019 with Ventura 13.4.1(c) which does have an Intel processor also. So it appears to be OS-related. I'll try to OCLP the trashcan to Ventura and see if this error disappears... would appreciate if the other folks with this error (and with natively Ventura-capable Macs) also test.
GiteaMirror commented 2026-04-28 02:31:17 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Nov 17, 2023):

Hi all, I'm so sorry you hit this bug on Intel Macs. This should be fixed as of 41434a7. A new release should be out early tomorrow but in the meantime you can try the prerelease version here https://github.com/jmorganca/ollama/releases/tag/v0.1.10. Again thanks for creating an issue and your patience!

<!-- gh-comment-id:1815685095 --> @jmorganca commented on GitHub (Nov 17, 2023): Hi all, I'm so sorry you hit this bug on Intel Macs. This should be fixed as of [`41434a7`](https://github.com/jmorganca/ollama/commit/41434a7cdcf33918ae2d37eb23d819ef7361e843). A new release should be out early tomorrow but in the meantime you can try the prerelease version here https://github.com/jmorganca/ollama/releases/tag/v0.1.10. Again thanks for creating an issue and your patience!
GiteaMirror commented 2026-04-28 02:31:19 -05:00
Author
Owner
Copy Link

@xyklex commented on GitHub (Dec 10, 2023):

Hi @jmorganca I've downloaded Ollama version 0.1.14, but it is running entirely on CPUs, is there a flag I should enable before running the model?
image

<!-- gh-comment-id:1849120696 --> @xyklex commented on GitHub (Dec 10, 2023): Hi @jmorganca I've downloaded Ollama version 0.1.14, but it is running entirely on CPUs, is there a flag I should enable before running the model? ![image](https://github.com/jmorganca/ollama/assets/483287/f246f308-31ea-4f21-ab8f-9a8fc2023c33)
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#46992
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?