github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15734] Ollama not working on Mac M5 #35792

New Issue
Open
opened 2026-04-22 20:28:11 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-22 20:28:11 -05:00
Owner
Copy Link

Originally created by @pranitmodi on GitHub (Apr 21, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15734

What is the issue?

Keep getting this error - 500 Internal Server Error: llama runner process has terminated: %!w()

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @pranitmodi on GitHub (Apr 21, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15734 ### What is the issue? Keep getting this error - 500 Internal Server Error: llama runner process has terminated: %!w(<nil>) ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-22 20:28:11 -05:00
GiteaMirror commented 2026-04-22 20:28:14 -05:00
Author
Owner
Copy Link

@cl93a commented on GitHub (Apr 21, 2026):

Same issue here, wanted to try out Gemma4 on MLX, but all models (not just MLX) fail. Ollama 0.21.0. Model fails to load with 500 error. Logs show:

ggml_metal_init: picking default device: Apple M5
signal arrived during cgo execution
github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_get_default_buffer_type(0x0)
fault 0x1926755b0
llama runner terminated: exit status 2

<!-- gh-comment-id:4291679769 --> @cl93a commented on GitHub (Apr 21, 2026): Same issue here, wanted to try out Gemma4 on MLX, but all models (not just MLX) fail. Ollama 0.21.0. Model fails to load with 500 error. Logs show: ggml_metal_init: picking default device: Apple M5 signal arrived during cgo execution github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_backend_get_default_buffer_type(0x0) fault 0x1926755b0 llama runner terminated: exit status 2
GiteaMirror commented 2026-04-22 20:28:15 -05:00
Author
Owner
Copy Link

@mverrilli commented on GitHub (Apr 21, 2026):

I am guessing #15581 will address this. Posted this to the wrong issue.

<!-- gh-comment-id:4292394924 --> @mverrilli commented on GitHub (Apr 21, 2026): ~~I am guessing #15581 will address this.~~ Posted this to the wrong issue.
GiteaMirror commented 2026-04-22 20:28:16 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Apr 22, 2026):

@cl93a can you verify this fails for you on MLX? What model and tag were you trying to run? So far, this seems to be a GGML specific defect.

<!-- gh-comment-id:4297865890 --> @dhiltgen commented on GitHub (Apr 22, 2026): @cl93a can you verify this fails for you on MLX? What model and tag were you trying to run? So far, this seems to be a GGML specific defect.
GiteaMirror commented 2026-04-22 20:28:17 -05:00
Author
Owner
Copy Link

@cl93a commented on GitHub (Apr 22, 2026):

@dhiltgen — Confirmed: MLX works, GGML does not on M5 (Ollama v0.21.1). Was confused that :latest did not auto pick mlx.

Tested:

gemma4:e2b-mlx-bf16 ✅ loads and runs fine
gemma4:e4b-mlx-bf16 ✅ loads and runs fine
gemma4:latest (Q4_K_M, GGML) ❌ crashes — even after a full ollama rm + re-pull, ruling out a corrupted download
gemma3:4b / gemma3:12b (GGML) ❌ same crash
All GGML failures hit ggml_metal_init: picking default device: Apple M5 → signal arrived during cgo execution → fault 0x1926755b0 → exit status 2. MLX is the only working backend on M5 right now it appears.

<!-- gh-comment-id:4298891093 --> @cl93a commented on GitHub (Apr 22, 2026): @dhiltgen — Confirmed: MLX works, GGML does not on M5 (Ollama v0.21.1). Was confused that :latest did not auto pick mlx. Tested: gemma4:e2b-mlx-bf16 ✅ loads and runs fine gemma4:e4b-mlx-bf16 ✅ loads and runs fine gemma4:latest (Q4_K_M, GGML) ❌ crashes — even after a full ollama rm + re-pull, ruling out a corrupted download gemma3:4b / gemma3:12b (GGML) ❌ same crash All GGML failures hit ggml_metal_init: picking default device: Apple M5 → signal arrived during cgo execution → fault 0x1926755b0 → exit status 2. MLX is the only working backend on M5 right now it appears.
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#35792
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?