github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13833] Server error when attempting image generation (M5 mac) #55570

New Issue
Closed
opened 2026-04-29 09:26:03 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-29 09:26:03 -05:00
Owner
Copy Link

Originally created by @jeastman on GitHub (Jan 22, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/13833

What is the issue?

Attempted to generate an image using the x/z-image-turbo model and received a server error

I am running Ollama 0.14.3 on M5 Macbook Pro (latest MacOS).

❯ ollama --version
ollama version is 0.14.3

When I attempt the image generation, the API returns HTTP 500.

❯ ollama run x/z-image-turbo "A cat with a hello sign"
Error: 500 Internal Server Error: Post "http://127.0.0.1:57107/completion": EOF

The logs show the following error:

MLX error: [metal::Device] Unable to load kernel affine_qmm_t_nax_bfloat16_t_gs_32_b_8_bm64_bn64_bk64_wm2_wn2_alN_true_batch_0

A discussion at llmstudio-ai leads me to believe this could be a M5 specific issue.

Relevant log output

time=2026-01-21T20:15:51.918-06:00 level=INFO source=routes.go:1629 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/jeastman/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]"
time=2026-01-21T20:15:51.957-06:00 level=INFO source=images.go:501 msg="total blobs: 3374"
time=2026-01-21T20:15:51.960-06:00 level=INFO source=images.go:508 msg="total unused blobs removed: 0"
time=2026-01-21T20:15:51.961-06:00 level=INFO source=routes.go:1682 msg="Listening on 127.0.0.1:11434 (version 0.14.3)"
time=2026-01-21T20:15:51.961-06:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-01-21T20:15:51.961-06:00 level=INFO source=server.go:429 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --port 56914"
time=2026-01-21T20:15:52.083-06:00 level=INFO source=types.go:42 msg="inference compute" id=0 filter_id=0 library=Metal compute=0.0 name=Metal description="Apple M5" libdirs="" driver=0.0 pci_id="" type=discrete total="25.0 GiB" available="25.0 GiB"
[GIN] 2026/01/21 - 20:15:52 | 200 |      27.041µs |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/01/21 - 20:15:52 | 200 |          50µs |       127.0.0.1 | HEAD     "/"
[GIN] 2026/01/21 - 20:15:52 | 200 |   66.450334ms |       127.0.0.1 | POST     "/api/show"
time=2026-01-21T20:15:52.426-06:00 level=INFO source=server.go:136 msg="starting image runner subprocess" exe=/Applications/Ollama.app/Contents/Resources/ollama model=x/z-image-turbo:latest port=56921
time=2026-01-21T20:15:52.456-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:78: INFO MLX library initialized"
time=2026-01-21T20:15:52.456-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:79: INFO starting image runner model=x/z-image-turbo:latest port=56921"
time=2026-01-21T20:15:52.460-06:00 level=INFO source=server.go:122 msg=image-runner msg="Loading Z-Image model from manifest: x/z-image-turbo:latest..."
time=2026-01-21T20:15:52.460-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:91: INFO detected model type type=ZImagePipeline"
time=2026-01-21T20:15:52.601-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading tokenizer... ✓"
time=2026-01-21T20:15:53.530-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading text encoder... ✓"
time=2026-01-21T20:15:53.531-06:00 level=INFO source=server.go:122 msg=image-runner msg="  (4.5 GB, peak 4.5 GB)"
time=2026-01-21T20:15:54.796-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading transformer... ✓"
time=2026-01-21T20:15:54.796-06:00 level=INFO source=server.go:122 msg=image-runner msg="  (11.7 GB, peak 11.7 GB)"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading conv_in... ✓"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading mid block... ✓"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading up blocks... ✓ [4 blocks]"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading conv_norm_out... ✓"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loading conv_out... ✓"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  (11.9 GB, peak 11.9 GB)"
time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg="  Loaded in 2.41s (11.9 GB VRAM)"
time=2026-01-21T20:15:54.869-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:54 runner.go:138: INFO image runner listening addr=127.0.0.1:56921"
time=2026-01-21T20:15:54.928-06:00 level=INFO source=server.go:207 msg="image runner is ready" port=56921
time=2026-01-21T20:15:54.946-06:00 level=INFO source=server.go:122 msg=image-runner msg="MLX error: [metal::Device] Unable to load kernel affine_qmm_t_nax_bfloat16_t_gs_32_b_8_bm64_bn64_bk64_wm2_wn2_alN_true_batch_0"
time=2026-01-21T20:15:54.946-06:00 level=INFO source=server.go:122 msg=image-runner msg=" at /Users/runner/work/ollama/ollama/build/_deps/mlx-c-src/mlx/c/transforms.cpp:73"
[GIN] 2026/01/21 - 20:15:55 | 500 |  2.722190167s |       127.0.0.1 | POST     "/api/generate"
time=2026-01-21T20:20:55.100-06:00 level=INFO source=server.go:305 msg="stopping image runner subprocess" pid=51208

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.14.3

Originally created by @jeastman on GitHub (Jan 22, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/13833 ### What is the issue? Attempted to generate an image using the x/z-image-turbo model and received a server error I am running Ollama 0.14.3 on **M5** Macbook Pro (latest MacOS). ``` ❯ ollama --version ollama version is 0.14.3 ``` When I attempt the image generation, the API returns HTTP 500. ``` ❯ ollama run x/z-image-turbo "A cat with a hello sign" Error: 500 Internal Server Error: Post "http://127.0.0.1:57107/completion": EOF ``` The logs show the following error: ``` MLX error: [metal::Device] Unable to load kernel affine_qmm_t_nax_bfloat16_t_gs_32_b_8_bm64_bn64_bk64_wm2_wn2_alN_true_batch_0 ``` A discussion at [llmstudio-ai](https://github.com/lmstudio-ai/lmstudio-bug-tracker/issues/1356) leads me to believe this could be a M5 specific issue. ### Relevant log output ```shell time=2026-01-21T20:15:51.918-06:00 level=INFO source=routes.go:1629 msg="server config" env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/jeastman/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]" time=2026-01-21T20:15:51.957-06:00 level=INFO source=images.go:501 msg="total blobs: 3374" time=2026-01-21T20:15:51.960-06:00 level=INFO source=images.go:508 msg="total unused blobs removed: 0" time=2026-01-21T20:15:51.961-06:00 level=INFO source=routes.go:1682 msg="Listening on 127.0.0.1:11434 (version 0.14.3)" time=2026-01-21T20:15:51.961-06:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-01-21T20:15:51.961-06:00 level=INFO source=server.go:429 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --port 56914" time=2026-01-21T20:15:52.083-06:00 level=INFO source=types.go:42 msg="inference compute" id=0 filter_id=0 library=Metal compute=0.0 name=Metal description="Apple M5" libdirs="" driver=0.0 pci_id="" type=discrete total="25.0 GiB" available="25.0 GiB" [GIN] 2026/01/21 - 20:15:52 | 200 | 27.041µs | 127.0.0.1 | GET "/api/version" [GIN] 2026/01/21 - 20:15:52 | 200 | 50µs | 127.0.0.1 | HEAD "/" [GIN] 2026/01/21 - 20:15:52 | 200 | 66.450334ms | 127.0.0.1 | POST "/api/show" time=2026-01-21T20:15:52.426-06:00 level=INFO source=server.go:136 msg="starting image runner subprocess" exe=/Applications/Ollama.app/Contents/Resources/ollama model=x/z-image-turbo:latest port=56921 time=2026-01-21T20:15:52.456-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:78: INFO MLX library initialized" time=2026-01-21T20:15:52.456-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:79: INFO starting image runner model=x/z-image-turbo:latest port=56921" time=2026-01-21T20:15:52.460-06:00 level=INFO source=server.go:122 msg=image-runner msg="Loading Z-Image model from manifest: x/z-image-turbo:latest..." time=2026-01-21T20:15:52.460-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:52 runner.go:91: INFO detected model type type=ZImagePipeline" time=2026-01-21T20:15:52.601-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading tokenizer... ✓" time=2026-01-21T20:15:53.530-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading text encoder... ✓" time=2026-01-21T20:15:53.531-06:00 level=INFO source=server.go:122 msg=image-runner msg=" (4.5 GB, peak 4.5 GB)" time=2026-01-21T20:15:54.796-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading transformer... ✓" time=2026-01-21T20:15:54.796-06:00 level=INFO source=server.go:122 msg=image-runner msg=" (11.7 GB, peak 11.7 GB)" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading conv_in... ✓" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading mid block... ✓" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading up blocks... ✓ [4 blocks]" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading conv_norm_out... ✓" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loading conv_out... ✓" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" (11.9 GB, peak 11.9 GB)" time=2026-01-21T20:15:54.869-06:00 level=INFO source=server.go:122 msg=image-runner msg=" Loaded in 2.41s (11.9 GB VRAM)" time=2026-01-21T20:15:54.869-06:00 level=WARN source=server.go:129 msg=image-runner msg="2026/01/21 20:15:54 runner.go:138: INFO image runner listening addr=127.0.0.1:56921" time=2026-01-21T20:15:54.928-06:00 level=INFO source=server.go:207 msg="image runner is ready" port=56921 time=2026-01-21T20:15:54.946-06:00 level=INFO source=server.go:122 msg=image-runner msg="MLX error: [metal::Device] Unable to load kernel affine_qmm_t_nax_bfloat16_t_gs_32_b_8_bm64_bn64_bk64_wm2_wn2_alN_true_batch_0" time=2026-01-21T20:15:54.946-06:00 level=INFO source=server.go:122 msg=image-runner msg=" at /Users/runner/work/ollama/ollama/build/_deps/mlx-c-src/mlx/c/transforms.cpp:73" [GIN] 2026/01/21 - 20:15:55 | 500 | 2.722190167s | 127.0.0.1 | POST "/api/generate" time=2026-01-21T20:20:55.100-06:00 level=INFO source=server.go:305 msg="stopping image runner subprocess" pid=51208 ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.14.3
GiteaMirror added the macosbug labels 2026-04-29 09:26:03 -05:00
GiteaMirror commented 2026-04-29 09:26:05 -05:00
Author
Owner
Copy Link

@kolargol commented on GitHub (Jan 24, 2026):

The same happens on ollama 0.15:

ollama --version
ollama version is 0.15.0

❯ ollama run x/flux2-klein:9b "/Users/zbyszek/man-flying-rocket-with-a-dog-rocket-is-made-of-che-20260115-202651.png replace dog with the shark"
Added image '/Users/zbyszek/man-flying-rocket-with-a-dog-rocket-is-made-of-che-20260115-202651.png'
Error: 500 Internal Server Error: model does not support image editing

i am pulling latest x/flux2-klein:9b

<!-- gh-comment-id:3794317245 --> @kolargol commented on GitHub (Jan 24, 2026): The same happens on ollama 0.15: ollama --version ollama version is 0.15.0 ❯ ollama run x/flux2-klein:9b "/Users/zbyszek/man-flying-rocket-with-a-dog-rocket-is-made-of-che-20260115-202651.png replace dog with the shark" Added image '/Users/zbyszek/man-flying-rocket-with-a-dog-rocket-is-made-of-che-20260115-202651.png' Error: 500 Internal Server Error: model does not support image editing i am pulling latest x/flux2-klein:9b
GiteaMirror commented 2026-04-29 09:26:05 -05:00
Author
Owner
Copy Link

@ridvan70 commented on GitHub (Jan 25, 2026):

Different but still error on llama 0.15.1

@Mac-M2 ~ % ollama --version
ollama version is 0.15.1
@Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist
@Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist

#################
NAME                              ID              SIZE      MODIFIED
x/flux2-klein:9b                  5fd79ad76b03    11 GB     3 seconds ago
x/flux2-klein:latest              8c7f37810489    5.7 GB    9 seconds ago
<!-- gh-comment-id:3796877319 --> @ridvan70 commented on GitHub (Jan 25, 2026): Different but still error on llama 0.15.1 ``` @Mac-M2 ~ % ollama --version ollama version is 0.15.1 @Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world" pulling manifest Error: pull model manifest: file does not exist @Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world" pulling manifest Error: pull model manifest: file does not exist ################# NAME ID SIZE MODIFIED x/flux2-klein:9b 5fd79ad76b03 11 GB 3 seconds ago x/flux2-klein:latest 8c7f37810489 5.7 GB 9 seconds ago ```
GiteaMirror commented 2026-04-29 09:26:06 -05:00
Author
Owner
Copy Link

@ed-norris commented on GitHub (Jan 27, 2026):

Different but still error on llama 0.15.1

@Mac-M2 ~ % ollama --version
ollama version is 0.15.1
@Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist
@Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist

#################
NAME                              ID              SIZE      MODIFIED
x/flux2-klein:9b                  5fd79ad76b03    11 GB     3 seconds ago
x/flux2-klein:latest              8c7f37810489    5.7 GB    9 seconds ago

Is that a typo - flux-klein instead of flux2-klein ?

<!-- gh-comment-id:3803096186 --> @ed-norris commented on GitHub (Jan 27, 2026): > Different but still error on llama 0.15.1 > > ``` > @Mac-M2 ~ % ollama --version > ollama version is 0.15.1 > @Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world" > pulling manifest > Error: pull model manifest: file does not exist > @Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world" > pulling manifest > Error: pull model manifest: file does not exist > > ################# > NAME ID SIZE MODIFIED > x/flux2-klein:9b 5fd79ad76b03 11 GB 3 seconds ago > x/flux2-klein:latest 8c7f37810489 5.7 GB 9 seconds ago > ``` Is that a typo - `flux-klein` instead of `flux2-klein` ?
GiteaMirror commented 2026-04-29 09:26:06 -05:00
Author
Owner
Copy Link

@ridvan70 commented on GitHub (Jan 27, 2026):

Different but still error on llama 0.15.1

@Mac-M2 ~ % ollama --version
ollama version is 0.15.1
@Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist
@Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world"
pulling manifest
Error: pull model manifest: file does not exist

#################
NAME                              ID              SIZE      MODIFIED
x/flux2-klein:9b                  5fd79ad76b03    11 GB     3 seconds ago
x/flux2-klein:latest              8c7f37810489    5.7 GB    9 seconds ago

Is that a typo - flux-klein instead of flux2-klein ?

Fine Observation, thanks. I copy/pasted it from Ollama page and did not verify it. I will correct and retry: Done, it works

<!-- gh-comment-id:3806887730 --> @ridvan70 commented on GitHub (Jan 27, 2026): > > Different but still error on llama 0.15.1 > > ``` > > @Mac-M2 ~ % ollama --version > > ollama version is 0.15.1 > > @Mac-M2 ~ % ollama run x/flux-klein "a cat holding a sign that says hello world" > > pulling manifest > > Error: pull model manifest: file does not exist > > @Mac-M2 ~ % ollama run x/flux-klein:9b "a cat holding a sign that says hello world" > > pulling manifest > > Error: pull model manifest: file does not exist > > > > ################# > > NAME ID SIZE MODIFIED > > x/flux2-klein:9b 5fd79ad76b03 11 GB 3 seconds ago > > x/flux2-klein:latest 8c7f37810489 5.7 GB 9 seconds ago > > ``` > > Is that a typo - `flux-klein` instead of `flux2-klein` ? Fine Observation, thanks. I copy/pasted it from Ollama page and did not verify it. I will correct and retry: Done, it works
GiteaMirror commented 2026-04-29 09:26:06 -05:00
Author
Owner
Copy Link

@jeastman commented on GitHub (Mar 3, 2026):

I tested this again today and it appears to now be working.

ollama --version
ollama version is 0.17.5

Thank you!

Closing the ticket.

<!-- gh-comment-id:3988598401 --> @jeastman commented on GitHub (Mar 3, 2026): I tested this again today and it appears to now be working. ```shell ollama --version ollama version is 0.17.5 ``` Thank you! Closing the ticket.
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 macos
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#55570
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?