github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-05 23:53:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7570] How to install Olama in a distributed manner #51334

New Issue
Closed
opened 2026-04-28 19:33:12 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-28 19:33:12 -05:00
Owner
Copy Link

Originally created by @smileyboy2019 on GitHub (Nov 8, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7570

How to connect two servers with 4090 graphics cards and provide unified services

Originally created by @smileyboy2019 on GitHub (Nov 8, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7570 How to connect two servers with 4090 graphics cards and provide unified services
GiteaMirror added the feature request label 2026-04-28 19:33:12 -05:00
GiteaMirror commented 2026-04-28 19:33:36 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 8, 2024):

ollama doesn't currently support distributed inference. See #6729 for ongoing work.

If you don't need to run a single model across multiple distributed cards, then a load balancing proxy like litellm or ollama_proxy might work for you.

<!-- gh-comment-id:2464459557 --> @rick-github commented on GitHub (Nov 8, 2024): ollama doesn't currently support distributed inference. See #6729 for ongoing work. If you don't need to run a single model across multiple distributed cards, then a load balancing proxy like [litellm](https://github.com/BerriAI/litellm) or [ollama_proxy](https://github.com/ParisNeo/ollama_proxy_server) might work for you.
GiteaMirror commented 2026-04-28 19:33:37 -05:00
Author
Owner
Copy Link

@Jotschi commented on GitHub (Nov 8, 2024):

I use nginx for this. No need for a specific service. The clusters can be selected using the "X-LLM-Cluster" custom http header. I use this to select different ollama clusters via a single endpoint.


upstream ollama_nemo {
    least_conn;

    server SERVER_1:11436 max_conns=8; # GPU 1
    server SERVER_2:11436 max_conns=8; # GPU 2
}

upstream ollama_fallback {
    least_conn;

    server SERVER_1:11439 max_conns=8; # GPU 1
}

server {
    listen       8080;
    server_name  localhost;

    location / {
        if ($http_x_llm_cluster = "nemo") {
            proxy_pass http://ollama_nemo;
        }
        
        proxy_pass http://ollama_fallback;
        #return 400 'X-LLM-Cluster header not found or invalid';
    }

    # redirect server error pages to the static page /50x.html
    error_page   500 502 503 504  /50x.html;
    location = /50x.html {
        root   /usr/share/nginx/html;
    }
}
<!-- gh-comment-id:2464469733 --> @Jotschi commented on GitHub (Nov 8, 2024): I use nginx for this. No need for a specific service. The clusters can be selected using the "X-LLM-Cluster" custom http header. I use this to select different ollama clusters via a single endpoint. ``` upstream ollama_nemo { least_conn; server SERVER_1:11436 max_conns=8; # GPU 1 server SERVER_2:11436 max_conns=8; # GPU 2 } upstream ollama_fallback { least_conn; server SERVER_1:11439 max_conns=8; # GPU 1 } server { listen 8080; server_name localhost; location / { if ($http_x_llm_cluster = "nemo") { proxy_pass http://ollama_nemo; } proxy_pass http://ollama_fallback; #return 400 'X-LLM-Cluster header not found or invalid'; } # redirect server error pages to the static page /50x.html error_page 500 502 503 504 /50x.html; location = /50x.html { root /usr/share/nginx/html; } } ```
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 feature request
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#51334
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?