github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 08:59:13 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15110] M1 Mac: z-image-turbo:fp8 VRAM regression 0.1.8.1→0.1.8.2 & 0.1.8.3 ... (11.9>11.3 GiB) #9680

New Issue
Open
opened 2026-04-12 22:33:53 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 22:33:53 -05:00
Owner
Copy Link

Originally created by @Toubinator on GitHub (Mar 27, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15110

Was ist das Problem?

x/z-image-turbo:fp8 lief stabil auf v0.1.8.1 (M1 Mac Mini 16GB).
Nach Update: "model requires 11.9 GiB but only 11.3 GiB available (512 MiB overhead)" – auch bei 512x512.

Erwartetes Verhalten

Modell lädt wie vorher (flexible M1 unified memory).

Betriebssystem

macOS Sonoma/Ventura (M1 Mac Mini 16GB unified memory)

GPU

Apple M1 (16GB)

Schritte zur Reproduktion

  1. ollama run x/z-image-turbo:fp8 "test" --width 512 --height 512
  2. Error trotz kleiner Auflösung

Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead)

Zusätzlich

  • v0.1.8.1: Funktioniert
  • Nach Update: Fail
Originally created by @Toubinator on GitHub (Mar 27, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15110 ### Was ist das Problem? x/z-image-turbo:fp8 lief stabil auf v0.1.8.1 (M1 Mac Mini 16GB). Nach Update: "model requires 11.9 GiB but only 11.3 GiB available (512 MiB overhead)" – auch bei 512x512. ### Erwartetes Verhalten Modell lädt wie vorher (flexible M1 unified memory). ### Betriebssystem macOS Sonoma/Ventura (M1 Mac Mini 16GB unified memory) ### GPU Apple M1 (16GB) ### Schritte zur Reproduktion 1. `ollama run x/z-image-turbo:fp8 "test" --width 512 --height 512` 2. Error trotz kleiner Auflösung Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead) ### Zusätzlich - v0.1.8.1: Funktioniert - Nach Update: Fail
GiteaMirror added the imagememorymacos labels 2026-04-12 22:33:53 -05:00
GiteaMirror commented 2026-04-12 22:33:54 -05:00
Author
Owner
Copy Link

@Toubinator commented on GitHub (Mar 28, 2026):

(Englisch) version

What is the problem?

x/z-image-turbo:fp8 ran stably on v0.1.8.1 (M1 Mac Mini 16GB).
After update: Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead) – even with 512x512 resolution.

Expected Behavior

Model loads as before (flexible M1 unified memory handling).

Ollama Version

v0.1.8.1 (working) vs latest (broken)

Operating System

macOS Tahoe 26.3.1 (a) (25D771280a)
Apple M1 Mac Mini 16GB unified memory

GPU

Apple M1 (16GB unified memory)

Steps to reproduce

ollama run x/z-image-turbo:fp8 "test cat" --width 512 --height 512 --steps 12
Error despite minimal resolution.

Error Log

Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead)

Additional Context

  • v0.1.8.1: ✅ Works perfectly
  • After update: ❌ Fails consistently
  • RAM free: ~12GB before run
  • ollama ps: Empty before loading
  • Downgrade to v0.1.8.1: ✅ Immediate fix
  • Alternative x/flux2-klein:latest: ✅ Works (5.7GB)

Environment

  • Native Ollama (no Docker)
  • M1 Mac Mini 16GB
  • macOS 26.3.1 (a) (25D771280a)
<!-- gh-comment-id:4148446312 --> @Toubinator commented on GitHub (Mar 28, 2026): (Englisch) version ### What is the problem? x/z-image-turbo:fp8 ran stably on v0.1.8.1 (M1 Mac Mini 16GB). After update: `Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead)` – even with 512x512 resolution. ### Expected Behavior Model loads as before (flexible M1 unified memory handling). ### Ollama Version v0.1.8.1 (working) vs latest (broken) ### Operating System macOS Tahoe 26.3.1 (a) (25D771280a) Apple M1 Mac Mini 16GB unified memory ### GPU Apple M1 (16GB unified memory) ### Steps to reproduce ollama run x/z-image-turbo:fp8 "test cat" --width 512 --height 512 --steps 12 Error despite minimal resolution. ### Error Log Error: 500 Internal Server Error: model requires 11.9 GiB but only 11.3 GiB are available (after 512.0 MiB overhead) ### Additional Context - **v0.1.8.1**: ✅ Works perfectly - **After update**: ❌ Fails consistently - **RAM free**: ~12GB before run - **`ollama ps`**: Empty before loading - **Downgrade to v0.1.8.1**: ✅ Immediate fix - **Alternative `x/flux2-klein:latest`**: ✅ Works (5.7GB) ### Environment - Native Ollama (no Docker) - M1 Mac Mini 16GB - macOS 26.3.1 (a) (25D771280a)
GiteaMirror commented 2026-04-12 22:33:55 -05:00
Author
Owner
Copy Link

@AbdulrahmanKanaan commented on GitHub (Apr 4, 2026):

I have the same issue; it was working fine on older versions.

<!-- gh-comment-id:4186904271 --> @AbdulrahmanKanaan commented on GitHub (Apr 4, 2026): I have the same issue; it was working fine on older versions.
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 image macos memory
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#9680
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?