github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4747] Running multiple models simultaneously, always using one card #2990

New Issue
Closed
opened 2026-04-12 13:23:04 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 13:23:04 -05:00
Owner
Copy Link

Originally created by @leoHostProject on GitHub (May 31, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4747

What is the issue?

Running multiple models simultaneously, always using one card,but i have 4 cards and download 4model
When multiple users are using it at the same time, always clear the first card and then load other models instead of using my other idle cards

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.33

Originally created by @leoHostProject on GitHub (May 31, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4747 ### What is the issue? Running multiple models simultaneously, always using one card,but i have 4 cards and download 4model When multiple users are using it at the same time, always clear the first card and then load other models instead of using my other idle cards ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.33
GiteaMirror added the bug label 2026-04-12 13:23:04 -05:00
GiteaMirror commented 2026-04-12 13:23:08 -05:00
Author
Owner
Copy Link

@leoHostProject commented on GitHub (May 31, 2024):

this is my ollama settings
Execstart=/usr/local/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
Environment="PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
Environment="OLLAMA HOST=0.0.0.0"
Environment="CUDA VISIBLE DEVICES=0,1,2,3"
Environment="OLLAMA NUM PARALLEL=4"

<!-- gh-comment-id:2141433906 --> @leoHostProject commented on GitHub (May 31, 2024): this is my ollama settings Execstart=/usr/local/bin/ollama serve User=ollama Group=ollama Restart=always RestartSec=3 Environment="PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin" Environment="OLLAMA HOST=0.0.0.0" Environment="CUDA VISIBLE DEVICES=0,1,2,3" Environment="OLLAMA NUM PARALLEL=4"
GiteaMirror commented 2026-04-12 13:23:08 -05:00
Author
Owner
Copy Link

@leoHostProject commented on GitHub (May 31, 2024):

gpu resources is NVIDIA A100 80GB PCIe *4
NVIDIA-SMI 535.154.05
CUDA Version:12.2
GPU Fan is 0,1,2,3
alwalys use 0 free 1,2,3

<!-- gh-comment-id:2141442707 --> @leoHostProject commented on GitHub (May 31, 2024): gpu resources is NVIDIA A100 80GB PCIe *4 NVIDIA-SMI 535.154.05 CUDA Version:12.2 GPU Fan is 0,1,2,3 alwalys use 0 free 1,2,3
GiteaMirror commented 2026-04-12 13:23:09 -05:00
Author
Owner
Copy Link

@DuckyBlender commented on GitHub (May 31, 2024):

please update your ollama to see if it's been fixed in the newer versions

<!-- gh-comment-id:2141572160 --> @DuckyBlender commented on GitHub (May 31, 2024): please update your ollama to see if it's been fixed in the newer versions
GiteaMirror commented 2026-04-12 13:23:10 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 31, 2024):

@leoHostProject to load 4 models on 4 GPUs concurrently, you need to also set OLLAMA_MAX_LOADED_MODELS - the current default is 1. In a future version this will be autodetected, but for now while it's still experimental, you have to set this variable to allow concurrent model loading.

<!-- gh-comment-id:2142821584 --> @dhiltgen commented on GitHub (May 31, 2024): @leoHostProject to load 4 models on 4 GPUs concurrently, you need to also set `OLLAMA_MAX_LOADED_MODELS` - the current default is 1. In a future version this will be autodetected, but for now while it's still experimental, you have to set this variable to allow concurrent model loading.
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#2990
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?