github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #16044] nvfp4 models stopped working on Apple M3 #87900

New Issue
Open
opened 2026-05-10 06:33:51 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-10 06:33:51 -05:00
Owner
Copy Link

Originally created by @johnkea on GitHub (May 8, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/16044

What is the issue?

A recent update decided to wage a personal vendetta against my local Ollama setup with *-nvfp4 - Gemma4 in particular. The charming error message reads: "This library is using a deployment target (0x00020008) that is not supported on this OS." Very helpful, truly.

I'm not upset. I'm fine. The installer, in its infinite wisdom, chose the silent treatment - not a single word about breaking changes, just a casual drive by sabotage. No warning, no changelog pop-up, no "hey buddy, this might ruin your day." Just vibes and a broken setup.

Relevant log output

% ollama run gemma4:26b-nvfp4
Error: 500 Internal Server Error: mlx runner failed: panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found  at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73 (exit: exit status 2)

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.23.2

Originally created by @johnkea on GitHub (May 8, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/16044 ### What is the issue? A recent update decided to wage a personal vendetta against my local Ollama setup with *-nvfp4 - Gemma4 in particular. The charming error message reads: "This library is using a deployment target (0x00020008) that is not supported on this OS." Very helpful, truly. I'm not upset. I'm fine. The installer, in its infinite wisdom, chose the silent treatment - not a single word about breaking changes, just a casual drive by sabotage. No warning, no changelog pop-up, no "hey buddy, this might ruin your day." Just vibes and a broken setup. ### Relevant log output ```shell % ollama run gemma4:26b-nvfp4 Error: 500 Internal Server Error: mlx runner failed: panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73 (exit: exit status 2) ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.23.2
GiteaMirror added the bugneeds more info labels 2026-05-10 06:33:51 -05:00
GiteaMirror commented 2026-05-10 06:33:52 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (May 8, 2026):

@johnkea can you post the logs? They should be in ~/.ollama/logs/server.log.

<!-- gh-comment-id:4409625108 --> @pdevine commented on GitHub (May 8, 2026): @johnkea can you post the logs? They should be in `~/.ollama/logs/server.log`.
GiteaMirror commented 2026-05-10 06:33:53 -05:00
Author
Owner
Copy Link

@johnkea commented on GitHub (May 9, 2026):

@johnkea can you post the logs? They should be in ~/.ollama/logs/server.log.

time=2026-05-09T11:42:41.192+02:00 level=INFO source=model_recommendations.go:177 msg="model recommendations cache sleep scheduled" wait=4h9m18.299392913s consecutive_failures=0
[GIN] 2026/05/09 - 12:40:05 | 200 | 58.875µs | 127.0.0.1 | HEAD "/"
[GIN] 2026/05/09 - 12:40:05 | 200 | 42.856208ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/05/09 - 12:40:05 | 200 | 24.52825ms | 127.0.0.1 | POST "/api/show"
time=2026-05-09T12:40:05.527+02:00 level=INFO source=sched.go:484 msg="system memory" total="36.0 GiB" free="25.4 GiB" free_swap="0 B"
time=2026-05-09T12:40:05.527+02:00 level=INFO source=sched.go:491 msg="gpu memory" id=0 library=Metal available="26.5 GiB" free="27.0 GiB" minimum="512.0 MiB" overhead="0 B"
time=2026-05-09T12:40:05.527+02:00 level=INFO source=client.go:359 msg="starting mlx runner subprocess" model=gemma4:26b-nvfp4 port=63712
time=2026-05-09T12:40:05.529+02:00 level=INFO source=sched.go:561 msg="loaded runners" count=1
time=2026-05-09T12:40:05.539+02:00 level=INFO source=server.go:44 msg="MLX engine initialized" "MLX version"=0.31.2-7-ge8ebdeb device=gpu
time=2026-05-09T12:40:05.609+02:00 level=INFO source=base.go:110 msg="Model architecture" arch=Gemma4ForConditionalGeneration
time=2026-05-09T12:40:06.183+02:00 level=INFO source=runner.go:159 msg="Loaded tensors from manifest" count=1498
panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73

goroutine 1 [running]:
github.com/ollama/ollama/x/internal/mlxthread.(*Thread).Do(0x1036c603b?, {0x10461e3a0?, 0x104869360?}, 0x1036b7b00?)
/Users/runner/work/ollama/ollama/x/internal/mlxthread/thread.go:64 +0x60
github.com/ollama/ollama/x/mlxrunner.Execute({0x27197a0a8110, 0x4, 0x4})
/Users/runner/work/ollama/ollama/x/mlxrunner/server.go:64 +0x428
github.com/ollama/ollama/runner.Execute({0x27197a0a80f0?, 0x0?, 0x0?})
/Users/runner/work/ollama/ollama/runner/runner.go:22 +0xc4
github.com/ollama/ollama/cmd.NewCLI.func3(0x27197a58f000?, {0x1036b7908?, 0x4?, 0x1036b790c?})
/Users/runner/work/ollama/ollama/cmd/cmd.go:2440 +0x50
github.com/spf13/cobra.(*Command).execute(0x27197a391b08, {0x27197a408820, 0x5, 0x5})
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x5f8
github.com/spf13/cobra.(*Command).ExecuteC(0x27197a732f08)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x2ec
github.com/spf13/cobra.(*Command).Execute(...)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
/Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
/Users/runner/work/ollama/ollama/main.go:12 +0x54
time=2026-05-09T12:40:06.190+02:00 level=ERROR source=sched.go:567 msg="error loading llama server" error="mlx runner failed: panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73 (exit: exit status 2)"
time=2026-05-09T12:40:06.190+02:00 level=INFO source=client.go:126 msg="stopping mlx runner subprocess" pid=91289
[GIN] 2026/05/09 - 12:40:06 | 500 | 703.887083ms | 127.0.0.1 | POST "/api/generate"

<!-- gh-comment-id:4412338003 --> @johnkea commented on GitHub (May 9, 2026): > [@johnkea](https://github.com/johnkea) can you post the logs? They should be in `~/.ollama/logs/server.log`. time=2026-05-09T11:42:41.192+02:00 level=INFO source=model_recommendations.go:177 msg="model recommendations cache sleep scheduled" wait=4h9m18.299392913s consecutive_failures=0 [GIN] 2026/05/09 - 12:40:05 | 200 | 58.875µs | 127.0.0.1 | HEAD "/" [GIN] 2026/05/09 - 12:40:05 | 200 | 42.856208ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/05/09 - 12:40:05 | 200 | 24.52825ms | 127.0.0.1 | POST "/api/show" time=2026-05-09T12:40:05.527+02:00 level=INFO source=sched.go:484 msg="system memory" total="36.0 GiB" free="25.4 GiB" free_swap="0 B" time=2026-05-09T12:40:05.527+02:00 level=INFO source=sched.go:491 msg="gpu memory" id=0 library=Metal available="26.5 GiB" free="27.0 GiB" minimum="512.0 MiB" overhead="0 B" time=2026-05-09T12:40:05.527+02:00 level=INFO source=client.go:359 msg="starting mlx runner subprocess" model=gemma4:26b-nvfp4 port=63712 time=2026-05-09T12:40:05.529+02:00 level=INFO source=sched.go:561 msg="loaded runners" count=1 time=2026-05-09T12:40:05.539+02:00 level=INFO source=server.go:44 msg="MLX engine initialized" "MLX version"=0.31.2-7-ge8ebdeb device=gpu time=2026-05-09T12:40:05.609+02:00 level=INFO source=base.go:110 msg="Model architecture" arch=Gemma4ForConditionalGeneration time=2026-05-09T12:40:06.183+02:00 level=INFO source=runner.go:159 msg="Loaded tensors from manifest" count=1498 panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73 goroutine 1 [running]: github.com/ollama/ollama/x/internal/mlxthread.(*Thread).Do(0x1036c603b?, {0x10461e3a0?, 0x104869360?}, 0x1036b7b00?) /Users/runner/work/ollama/ollama/x/internal/mlxthread/thread.go:64 +0x60 github.com/ollama/ollama/x/mlxrunner.Execute({0x27197a0a8110, 0x4, 0x4}) /Users/runner/work/ollama/ollama/x/mlxrunner/server.go:64 +0x428 github.com/ollama/ollama/runner.Execute({0x27197a0a80f0?, 0x0?, 0x0?}) /Users/runner/work/ollama/ollama/runner/runner.go:22 +0xc4 github.com/ollama/ollama/cmd.NewCLI.func3(0x27197a58f000?, {0x1036b7908?, 0x4?, 0x1036b790c?}) /Users/runner/work/ollama/ollama/cmd/cmd.go:2440 +0x50 github.com/spf13/cobra.(*Command).execute(0x27197a391b08, {0x27197a408820, 0x5, 0x5}) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:940 +0x5f8 github.com/spf13/cobra.(*Command).ExecuteC(0x27197a732f08) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:1068 +0x2ec github.com/spf13/cobra.(*Command).Execute(...) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) /Users/runner/go/pkg/mod/github.com/spf13/cobra@v1.7.0/command.go:985 main.main() /Users/runner/work/ollama/ollama/main.go:12 +0x54 time=2026-05-09T12:40:06.190+02:00 level=ERROR source=sched.go:567 msg="error loading llama server" error="mlx runner failed: panic: mlx: Failed to load the default metallib. This library is using a deployment target (0x00020008) that is not supported on this OS. library not found library not found library not found at /Users/runner/work/ollama/ollama/build/metal-v3/_deps/mlx-c-src/mlx/c/transforms.cpp:73 (exit: exit status 2)" time=2026-05-09T12:40:06.190+02:00 level=INFO source=client.go:126 msg="stopping mlx runner subprocess" pid=91289 [GIN] 2026/05/09 - 12:40:06 | 500 | 703.887083ms | 127.0.0.1 | POST "/api/generate"
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 needs more info
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#87900
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?