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 #2191] If you have multiple GPUs then the new default split_mode = "layer" option in the wrapped llama.cpp server may effect you alot! #63290

New Issue
Closed
opened 2026-05-03 12:52:48 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-03 12:52:48 -05:00
Owner
Copy Link

Originally created by @jukofyork on GitHub (Jan 25, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2191

I'm not sure why the llama.cpp devs have made the new default split_mode = "layer", but it runs MUCH worse for me and I only get around 60% of the tokens/s that I get with split_mode = "row" (using 2x RTX A6000 and an NvLink bridge).

The only difference I can see is that using split_mode = "layer" seems to allocate the VRAM much more evenly.

I'm also seeing nothing move through the NvLink bridge now unless I set main_gpu = 0, but this doesn't really have any effect on the poor performance of split_mode = "layer".

See https://github.com/ollama/ollama/pull/2190 for more details and the fix.

Originally created by @jukofyork on GitHub (Jan 25, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2191 I'm not sure why the `llama.cpp` devs have made the new default `split_mode = "layer"`, but it runs ***MUCH*** worse for me and I only get around 60% of the tokens/s that I get with `split_mode = "row"` (using 2x RTX A6000 and an NvLink bridge). The only difference I can see is that using `split_mode = "layer"` seems to allocate the VRAM much more evenly. I'm also seeing nothing move through the NvLink bridge now unless I set `main_gpu = 0`, but this doesn't really have any effect on the poor performance of `split_mode = "layer"`. See https://github.com/ollama/ollama/pull/2190 for more details and the fix.
GiteaMirror added the performancenvidia labels 2026-05-03 12:52:50 -05:00
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 nvidia performance
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#63290
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?