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 #10705] Error: llama runner process has terminated: GGML_ASSERT(n_backends <= GGML_SCHED_MAX_BACKENDS) failed #53547

New Issue
Closed
opened 2026-04-29 03:45:43 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-29 03:45:43 -05:00
Owner
Copy Link

Originally created by @fahadshery on GitHub (May 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10705

What is the issue?

Hi,

I just found the similar issue here...https://github.com/ollama/ollama/issues/8920

I am using 16 Nvidia A16 GPUs...

I am using Ollama in Docker....

Where do I set GGML_SCHED_MAX_BACKENDS?

thanks

Relevant log output

# ollama run llama4
Error: llama runner process has terminated: GGML_ASSERT(n_backends <= GGML_SCHED_MAX_BACKENDS) failed
# nvidia-smi
Wed May 14 07:21:23 2025       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.133.20             Driver Version: 570.133.20     CUDA Version: 12.8     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA A16                     Off |   00000000:1B:00.0 Off |                    0 |
|  0%   38C    P0             25W /   62W |      62MiB /  15356MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   1  NVIDIA A16                     Off |   00000000:1C:00.0 Off |                    0 |
|  0%   31C    P8             14W /   62W |       3MiB /  15356MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

snipped....

OS

Docker

GPU

16 x Nvidia A16

CPU

No response

Ollama version

0.6.8

Originally created by @fahadshery on GitHub (May 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10705 ### What is the issue? Hi, I just found the similar issue here...https://github.com/ollama/ollama/issues/8920 I am using 16 `Nvidia A16` GPUs... I am using Ollama in Docker.... Where do I set `GGML_SCHED_MAX_BACKENDS`? thanks ### Relevant log output ```shell # ollama run llama4 Error: llama runner process has terminated: GGML_ASSERT(n_backends <= GGML_SCHED_MAX_BACKENDS) failed # nvidia-smi Wed May 14 07:21:23 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 570.133.20 Driver Version: 570.133.20 CUDA Version: 12.8 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA A16 Off | 00000000:1B:00.0 Off | 0 | | 0% 38C P0 25W / 62W | 62MiB / 15356MiB | 0% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ | 1 NVIDIA A16 Off | 00000000:1C:00.0 Off | 0 | | 0% 31C P8 14W / 62W | 3MiB / 15356MiB | 0% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ snipped.... ``` ### OS Docker ### GPU 16 x Nvidia A16 ### CPU _No response_ ### Ollama version 0.6.8
GiteaMirror added the bug label 2026-04-29 03:45:43 -05:00
GiteaMirror commented 2026-04-29 03:45:45 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (May 14, 2025):

0aa8b371dd/ml/backend/ggml/ggml/src/ggml-backend.cpp (L611-L613)

<!-- gh-comment-id:2879354538 --> @rick-github commented on GitHub (May 14, 2025): https://github.com/ollama/ollama/blob/0aa8b371ddd24a2d0ce859903a9284e9544f5c78/ml/backend/ggml/ggml/src/ggml-backend.cpp#L611-L613
GiteaMirror commented 2026-04-29 03:45:45 -05:00
Author
Owner
Copy Link

@fahadshery commented on GitHub (May 14, 2025):

ggml-backend.cpp

Hi Rick,

where is the location of this file? I am using docker. is this something we could pass as env vars?

<!-- gh-comment-id:2880468347 --> @fahadshery commented on GitHub (May 14, 2025): > ggml-backend.cpp Hi Rick, where is the location of this file? I am using docker. is this something we could pass as `env` vars?
GiteaMirror commented 2026-04-29 03:45:46 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (May 14, 2025):

The location of the file is the link in the code snippet. This is a compile time option, it can't be influenced by environment variables. If you want to change it, clone the repo, edit the file, and build the image container.

<!-- gh-comment-id:2880487376 --> @rick-github commented on GitHub (May 14, 2025): The location of the file is the link in the code snippet. This is a compile time option, it can't be influenced by environment variables. If you want to change it, clone the repo, edit the file, and [build](https://github.com/ollama/ollama/blob/main/docs/development.md#docker) the image container.
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#53547
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?